Expression of Concern: Abstracts - 2019 - Basic &amp...

来自 : 发布时间：2025-04-04

Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat 001 Relationship between air pollution and outpatient visits for diabetes mellitus type 2Yi-Horng LaiDepartment of Health Care Administration, Oriental Institute of Technology, New Taipei City, TaiwanObjectives: Diabetes mellitus is a chronic disease that has been recognized as an important health issue in many countries. The prevalence rate of diabetes is very high in Taiwan that poses a serious threat to the health of people. Diabetes mellitus is a group of metabolic disorders characterized by hyperglycemia resulting characterized by hyperglycemia resulting from a lack of insulin effect. According to pervious study, insulin secretion has a negative relationship with long term exposure to polluted air. The purpose of this study is to examine the association between air pollution and the occurrence of diabetes mellitus type 2 (DMT2) in Taiwan. Methods: This study investigates the impact of air pollution on DMT2 based on the data from outpatient visits. The research focuses on the significance of air pollution\'s impact on ocular health. Data for outpatient visits were obtained from the National Health Insurance Research Database of Taiwan. Ambient air monitoring data were obtained from the Taiwan Environmental Protection Administration air pollution monitoring stations. Six regulated air pollutants were included in the studies. This study used symbolic data analysis with Julia 0.6.4 to investigate the association between daily air pollutant concentrations and the number of outpatient visits for DMT2 as well as the relationship between climate factors and air pollutants. Results: The result of this study shows that the air pollutant, CO (t-value=2.730, P-value=0.008) and PM2.5 (t-value=2.292, P-value=0.025) is significantly related to DMT2. The more the increase of CO and PM2.5, the more the cases of DMT2 occur. As for the relationship between climate factors and air pollutants, it can be found that daily temperature is also positively correlated with CO and PM2.5. The air pollutant, CO and PM2.5 can increase chances of outpatient visits, suggesting that is a possible cause for DMT2. This study also shows that a favorable air-pollution-monitoring infrastructure combined with the establishment of public health data can provide valuable information for an examination of adverse health effects related to air pollution. Conclusions: This study combined and integrated air pollution and health insurance data to investigate associations between outpatient visits for DMT2 and air pollution levels. It has found that the air pollutant, CO and PM2.5 can increase chances of outpatient visits, suggesting that it is a possible cause for DMT2.5. Furthermore, this study has also shown that a favorable air-pollution-monitoring infrastructure combined with established public health data can provide valuable information for an examination of adverse health effects related to air pollution.002 D-ConvNet: Deep learning model for enhancement of brain MR images Kathiravan Srinivasan1, Vishal Sharma2, Dushantha Nalin K. Jayakody3, Durai Raj Vincent11School of Information Technology and Engineering. Vellore Institute of Technology, Vellore, India, E-mail: kathiravan.srinivasan@vit.ac.in; 2Department of Information Security Engineering, Soonchunhyang University-Asan Si, The Republic of Korea;3School of Computer Science & Robotics, National Research Tomsk Polytechnic University, Russia Objectives: In clinical and medical imaging, the magnetic resonance (MR) images obtained, generally do not have a very high resolution because of the several factors like patient\'s comfort, scanning time, scanning equipment limitations, long sampling times, and so on. However, the MR imaging is always favored by physicians as one of the most trusted modes for clinical pathology, disease diagnosis, and treatment. Therefore, the enhancement of low-resolution MR image to a high-resolution MR image is critical for precise and effective clinical diagnosis. Furthermore, single-image super-resolution is an inverse problem because of its ill-posed characteristics. This problem can be surpassed by using deep learning models such as deep convolutional neural networks (D-ConvNet). Methods: In this research, the D-ConvNet is employed for enhancing the resolution of the low-resolution MR image. The D-ConvNet model possesses a single deconvolution layer, several convolution layers and output of each layer is sent through a Leaky Rectified Linear Unit. The high-resolution MR image is obtained by from a sequence of feature maps acquired from the preceding layer and the coinciding patches are aggregated by deploying the averaging process. Mean Square Error (MSE) is employed as the loss function to train the proposed model. The stochastic gradient descent approach with a conventional backpropagation is utilized for minimalizing the loss function and is performed ‘til a convergence is attained. Results: The performance of the proposed D-ConvNet model is assessed using the peak signal-to-noise ratio (PSNR) metric. The loss function defined by the MSE is inversely proportional to the PSNR, which is for smaller values of MSE will yield higher values of PSNR. Also, minimizing the loss function (MSE) will lead to an improved and enhanced image of higher resolution. In the training part, 1200 brain MR high-resolution images were used as the dataset for training. The interpolation factor is set as two, and the patch size was set to 20×20. A five-layer fully connected D-ConvNet was built with the initial network settings having the values f1=9, f2=9, f3=3, f4=3, f1=9 and the number of filters in each layer was correspondingly assigned to be 1, 56, 28, 28, and 1. The learning rate was set as 0.00001 for all the layers. It was discovered for a sagittal plane T2-weighted MR image that the PSNR of the proposed approach is 29.3693dB that is superior to SRCNN model with PSNR=27.2038dB. Moreover, better results can be achieved by increasing the number of the intermediate layers. Conclusions: In this work, a five-layered D-ConvNet was proposed to improve the resolution of low-resolution MR images. Further, this work exhibits better performance, higher efficiency, and strength than the SRCNN approach for MR Image enhancement. 1Department of Industrial Design, Chaoyang University of Technology, Taichung, Taiwan (ROC); 2Department of Industrial Design, Chaoyang University of Technology, Taichung, Taiwan (ROC); 3School of Economics and Management, Shandong Yingcai University, Jinan, China Objectives: Cycling is a good rehabilitation exercise. The main purpose of this study is to explore different riding postures during rehabilitation, and to correlate the muscle activity of lower limbs with the EMG signal as a model to verify the construction of human software. Through the inverse kinetic analysis to obtain the muscle activity of the user and the force value of the joint, and to explore the relationship between the different cushion positions for the muscle activity and the joint force to provide the reference data of the strength of the user in the rehabilitation training. Methods: This research mainly includes the pedaling power measurement platform, human body size measuring instrument, EMG measurement analysis and human factors riding posture analysis model experimental equipment and software, and then through the biomechanical human body model and 3D human body and The bicycle model and the size of the person and the car can be parameterized and can be applied to different users to further find the most suitable position and rehabilitation effect. Finally, 8 muscle signals are output, and the EMG signals are rectified, smoothed, and normalized to obtain muscle activity data values. Results: In the simulation analysis of the riding posture, the seat height setting is based on Greg LeMond\'s research. The standard seat cushion is raised to the origin, and the adjustment range of the seat cushion is adjusted by the distance of 5cm from the origin, the back, the upper and the lower, and the adjustment interval is 1cm each time, and a total of 21 biomechanical simulation analyses are performed; According to the Knight\'s personalized pedaling speed and power output curve to input the specific speed and power parameters, the user or the doctor can clearly know the difference between the pedaling speed, the human body output power and the cushion position adjustment during rehabilitation training. The user and the doctor select the appropriate rehabilitation bicycle and a suitable riding position. The study also found that the user\'s seat cushion adjustment to the standard seat height position 2cm forward and 1.5cm down adjustment allows the user to have the lowest muscle activity output, lower muscle activity represents muscle fatigue time can be extended, high output efficiency the riding time can be extended. Conclusions: In this study, the size of the human body is measured and parameterized into a 3D model. The computer biomechanical calculation is used to verify the correctness of the model through the measurement of the bioelectrical force. The results of experiments and analysis can provide a reference for users to recover. Department of Biomedical Engineering, School of Medicine, Keimyung University, Daegu, Republic of Korea Objectives: Impedance is a resistance that interrupts current flow. Recently, it has been utilized as various bio signal analysis methods through bioelectrical resistance analysis. Most of the evaluation for skin damage are identified in the form shown, and a step-by-step classification criterion is applied to the damage. This indicates that there is a limit to accurate diagnosis of skin damage. The purpose of this study was to investigate the bio resistance of skin induced by pressure ulcer using impedance to determine whether bio resistance measurement can be used as a diagnostic criterion for skin damage. Methods: In the experimental animal rats, a constant pressure (120mmHg, 2hr) was applied to the skin to induce the pressure ulcer, and the impedance was measured. Impedance magnitude and phase were measured using a Precision LCR meter with a 100mV constant-voltage sign wave output signal with a frequency between 100Hz and 1MHz. Results: Impedance values of normal skin tissues were measured from 1.117 to 9.664 │Z│Ω, and values of 3.255 to 4.793│Z│Ω were obtained for the group with induced pressure ulcers. Impedance measurements showed lower values when the pressure ulcer was induced, and the intervals decreased as the frequency measurements increased. As a result of this study, it was found that when the skin damage caused by the pressure ulcer occurs, the measured value of impedance is lower than normal condition, and it is confirmed that the impedance interval was large at low frequency. Conclusions: Skin lesions have been diagnosed and treated by the experience of the diagnose. These results suggest that a scientific and quantitative method can be applied to diagnosis of skin damage, which is dependent on the existing visual system, and it can be applied to diagnosis of various skin diseases. As a limit of this study, only the diagnosis by the pressure ulcer induction was performed using the impedance for skin damage. Skin damage is due to occur in various forms, standardization work on skin diseases is necessary. In addition, personalized diagnostic criteria should be presented because impedance values have different values depending on individual characteristics. Synthesis and characterization of polymer stabilized-iron oxide magnetic nanoparticles for the delivery of doxorubicin Janina Carrera1, Ndumiso Vukile Mdlovu1, Kuen-Song Lin1, Tzu-Wei Chang1, Yun Chen2, Sikhumbuzo Charles Kunene1 1Department of Chemical Engineering and Materials Science/Environmental Technology Research Centre, Yuan Ze University, Taoyuan City, Taiwan2Department of Surgery, Far Eastern Memorial Hospital, New Taipei City, 220, Taiwan and Department of Chemical Engineering & Materials Science, Yuan Ze University, Taiwan Background: Cancer is the major cause of death around the globe recording millions of new cases yearly. Chemotherapy has played a big role in improving the survival rate of cancer’s patients, it works by inducing cell apoptosis. However, cancer treatments using chemotherapy are generally complicated for patients due to the numerous side effects that this entails. Many nanomedical studies probe the potential to specifically target metal nanoparticles to detect and controlled tumor cells. This technology allows scientists to detect any molecular change even when it occurs only in a small percentage of cells. Method: A pH sensitive multifunctional magnetic drug delivery system (DDS) consisting of an iron oxide nanoparticle (IONP) core and a polymer shell formed by Pluronic F127 poly (ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide)(PEO-PPO-PEO) block copolymer, was developed via a co-precipitation method. The DDS was used as a carrier for the doxorubicin (DOX) anti-cancer drug to treat neuroblastoma. Several analytic techniques such as XRD, HR-TEM, XANES, and EXAFS were used to characterize the nanoparticles. Results: Regardless of the differences in Pluronic concentration and the dimensions of iron oxide nanoparticles products, all the samples displayed negligible coercivity and remanence, suggesting their superparamagnetic properties. Nanoparticles exhibiting nearly spherical shape and uniformly dispersed with sizes of 10–30nm were prepared. XANES spectra of Fe atom in the different samples demonstrated an absorbance feature (Fe=7113eV), of 1s to 3d transition. The diffraction peak of the synthesized magnetic nanocarriers at 2θ=35.44° corresponded to that of magnetite (Fe3O4) (311), consistent with the Joint Committee on Powder Diffraction Standards (JCPDS) database. The results of MTT assay showed that the prepared nanocomposites did not exhibit any significant cytotoxicity on treated cells. Furthermore, UV-Vis spectra of F127–IONPs–DOX confirmed the successful conjugation of the drug onto F127–IONPs. The invitro drug release profile displayed that the drug release from the F127–IONPs nanocarrier was pH sensitive, showing faster DOX release rate under acidic than in neutral conditions. Conclusions: Pluronic F127 coated iron oxide nanocomposites encapsulated with DOX, is a promising carrier for DDS in cancer chemotherapy. Finally, a more comprehensive understanding of the fundamental connection between the structure, property optimization of the synthesis condition, and polymer coating is necessary for the development of a novel drug delivery system. The in vivo animal test will be conducted to investigate the effect of the nanocarrier coupled with the drug. Acknowledgements: The financial support of the Ministry of Science and Technology (MOST), Taiwan, R.O.C. (MOST 105-3113-E-008-003) and the Taiwan Synchrotron Radiation Research Center (NSRRC) group for their helps in the XANES/EXAFS experiments are greatly acknowledged. An advanced computed tomography pattern classification method based on rough set and wavelet packet transform algorithms to support diagnosis of pulmonary tuberculosis Ching-Hsue Cheng1, Tai-Liang Chen2, Chun-Yu Lin1 1Dept. of Information Management, National Yunlin University of Science & Technology, Yunlin, Taiwan (R.O.C.); 2Dept. of Digital Content Applications and Management, Wenzao Ursuline University of Languages, Kaohsiung, Taiwan (R.O.C.) Objectives: As Taiwan\'s air pollution is becoming more and more serious in these years, the patients with lung disease seem to increase rapidly. And the average number of medical imaging data produced for Taiwanese is more than that of other countries. Therefore, using computer-assisted automatic diagnostics, based on artificial intelligence (AI) image analysis, to help doctors diagnose their patients with lung disease is a major approach to reduce diagnostic errors and improve the medical treatment quality. The research objective of this paper is to propose an advanced computed tomography pattern classification method to promote diagnostic accuracy. Methods: The proposed advanced computed tomography (CT) pattern classification method consists of four main blocks as follows. (1) Image processing block: adjust the contrast of chest CT images according to the density between lung area and thoracic cavity. Separate the lung areas from the background and thoracic cavity by the proposed automated pulmonary segmentation method. (2) Reconstruction block: uses the wavelet packet transform (WP) based on singular value decomposition (SVD) to produce the singular values of lungs areas, and the values are applied to outline the lung areas. The classification results can be used to compare lungs areas reconstructed by SVD with those images without reconstruction. (3) Feature extraction block: two main works are contained in this block: (i) apply wavelet packet based on singular value decomposition (WP-SVD) to reconstruct the lungs areas; and (ii) compute the feature values of the lungs areas from wavelet packet coefficients and reduce image attributes. (4) Classification block: apply the rough sets (LEM2 algorithm) to generate understandable rules, based on discrete wavelet packet transform (DWPT), for classifying image datasets. Results: In this paper, two datasets, from the Lung Image Database Consortium (LIDC) and a regional teaching Hospital (RTH), are used to verify the proposed method. Three types of lung disease for the LIDC datasets are: nodule, non-nodule and inflammation, and two types for the RTH datasets are: nodule, non-nodule. The classification accuracy of the proposed method for the LIDC and RTH are 99.41%, and 98.80%, respectively, and are much better than the comparison methods, including Trees.J48 (87.42%, 87.00%), Naïve Bayes (83.18%, 62.90%), and Multilayer Perception (84.48%, 89.50%). Conclusions: In the experimental results, the proposed method performs outstandingly in classification accuracy for diagnosis of lung disease and can be applied in diagnosis support system in the near future. Acknowledgements: This work was sponsored by Ministry of Science and Technology, Taiwan, R.O.C. (Grant no. MOST 107-2221-E-160-001). Brain as a potential target for antisense oligonucleotide treatment of hypertensive disease. Experimental study on the hypertensive ISIAH rat strain A.S. Levina1,2, V.F. Zarytova1,2, M.N. Repkova1,2, O.L. Klimov1,3, A.A. Seryapina1,3, N.V. Shikina1,4, O.B. Shevelev1,3, A.L. Markel1,3 1Novosibirsk State University, Novosibirsk, Russia; 2Institute of Chemical Biology and Fundamental Medicine, Siberian Branch of RAS, Novosibirsk, Russia; 3Institute of Cytology and Genetics, Siberian Branch of RAS, Novosibirsk, Russia4Institute of Catalysis, Siberian Branch of RAS, Novosibirsk, Russia Objectives: Brain is directly involved in the pathogenesis of the hypertensive disease. The invivo metabolome study of the hypothalamus in the hypertensive ISIAH rats revealed distinct changes in metabolites related to the energy and excitatory brain processes. Significant changes in norepinephrine and dopamine concentrations in the hypothalamus and the brain stem of the ISIAH rats were also found. Then, continuous brain intraventricular infusion of the blocker of the angiotensin-converting enzyme (ACE) by implanted micropumps produced a sustained fall in blood pressure (BP). This prompted us to study the effect of antisense oligodeoxyribonucleotides (ODN) targeted to mRNA of the ACE gene (ACE-mRNA) on the BP in the ISIAH rats. Methods: To study the antisense ODN effect on BP in the hypertensive ISIAH rats, we synthesized the Si˜NH2·ODN nanocomposite, amino-containing silicon nanoparticles bearing oligodeoxyribonucleotide targeted to ACE-mRNA. Si˜NH2·ODN was administered to rats by inhalation (100μM for ODN), and BP was measured (tail-cuff method) two days before and 14days after inhalation. To study the distribution of the nanocomposite in the body, ODN was labeled with the Cy7.5 dye. The Si˜NH2·ODN(Cy7.5) nanocomposite (2μM for ODN) was inhaled by the SCID hairless mice. The body distribution was detected by the fluorescent emission computed tomography immediately after inhalation, then 24h, 72h, and two weeks after inhalation. Results: After inhalation, the BP of the ISIAH rats decreased on average by 20mm/Hg for two weeks. The use of a scrambled oligonucleotide in the Si˜NH2·ODN nanocomposite had no effect. The distribution of the inhaled nanocomposite in the mice body demonstrated its predominant accumulation in the gastrointestinal tract. However, some amount of the ODN (Cy7.5) penetrated the blood-brain barrier and was observed in the brain stem. The signal in the brain increased during 24h after inhalation, and then decreased 1.4 and 2.4 times after three and seven days, respectively. The full excretion of the labeled sample from the body occurred during two weeks. Conclusions: ● Inhalation of Si˜NH2·ODN nanocomposite by hypertensive ISIAH rats produced two-week long BP decrease. ●Inhaled Si˜NH2·ODN nanocomposite containing antisense ODN against ACE-mRNA may reach the brain structures in the rodents. ●The results indicate a potential role of the brain renin-angiotensin system in the pathogenesis of stress-induced hypertension simulated by the ISIAH rat strain. Acknowledgments: The work was supported by the grant no. 16-15-10073 from the Russian Scientific Foundation. The studies were implemented using the equipment of the Center for Genetic Resources of Laboratory Animals at ICG SB RAS, supported by the Ministry of Education and Science of Russia (Unique identifier of the project RFMEFI62117X0015). The novel computer-aided diagnosis system for breast ultrasound images based on frequency domain features Jong-Ha Lee Dept. of Biomedical Eng., Keimyung University, School of Medicine, Daegu, South Korea Objectives: Screening through mammography is one of the most often used diagnostic tools for breast tumor detection. Mammography, however, it is not effective for dense breasts, with high false negative rates. In this research, we propose a mass irregularity feature extraction in Frequency domain approach for ultrasound images that are able to model subtle shape differences between benign and malignant in the global aspect of view, thus improving CAD classification accuracy. Methods: The Fourier coefficients features called Fourier Descriptor were extracted from ultrasound images. To verify the performance of the proposed features for breast tumor classification, experiments were done using 5252 breast ultrasound images of invivo human subjects containing 2507 malignant cases. Results: Experimental results show that the breast tumor classification accuracy is enhanced by 9.42%, 10.41%, 11.01%, and 33.92%, than the previous features such as spatial gray level dependence (SGLD) feature, depth-width ratio feature, number of depressions feature, and orientation feature, respectively. Conclusions: The Fourier descriptors were proved their efficiencies in the CAD system compared to other time domain features. Acknowledgements: This research was supported by NRF-2017R1D1A1B04031182, Ministry of Trade, Industry & Energy (MOTIE, Korea) under Industrial Technology Innovation Program, No. 10063553, R0004840, R0006452, and KHIDI (HI17C2594). Dept. of Information Management, National Yunlin University of Science & Technology, Yunlin, Taiwan (R.O.C.) Objectives: This study expects before identifying tumor, trains all abdominal CT of the collected datasets for two classes: \"with tumor” and \"without tumor”. Then the proposed system can be applied for those images which really need to be recognized. The objective is decreasing the time-consuming work of specialists and systems and looking forward to improve whole diagnosis process of tumor detection. Methods: This study collected two abdominal CT images, the first dataset included abdominal CT images selected randomly from 10 patients and 60 JPG images were used totally. The second dataset consisted of 160 abdominal CT images selected randomly from 8 patients and 320 normal images from 16 patients (practically collected from the region hospital in central Taiwan). Total 480 JPG images were used in the experiment. And this study proposed a hybrid method, which combined semi-automatic segmentation of region of interest (using TVSeg algorithm) with feature filtering (using discrete wavelet packet transform) and supervised rough set theory to classify two collected datasets. The proposed method can make sure that input CT image need to be detected out the actual information of tumors. Results: After experiments, the results show that the proposed method is better than C4.5, REP, MLP, Naïve Bayes, and SVM classifiers under Scheffe comparison. And we find that the proposed method has the lowest standard deviation, that is, it is robust and stable. Furthermore, different filters have no significant difference in statistical test, hence, the proposed method employs the filter \"haar” and different classifiers to analyze their results. Conclusions: In general, abdominal CT is a robust and the most commonly used imaging technique for the inspection of liver tumors. An efficient computerized aid diagnosis system (CAD) can assists physicians and radiologists conveniently acquire valuable information. To improve the CAD, the proposed method can enhance diagnostic ability for identifying tumors that have an obvious edge and a high difference between the average gray level intensity of the tumor. i.e., the research results can provide some suggestions in the training of radiologists. Department of Intelligent Engineering Informatics for Human, Sangmyung University, Seoul, South Korea Objectives: As a method for estimating disease and promoting health, a method of utilizing the constant monitoring information of a patient\'s condition by using a sensor that acquires various kinds of biometric information is widely used. Among them, respiration information is used to analyze the psychological state of the patient for the prevention of respiratory disorder during sleep and the psychiatric trauma treatment. However, the sensor that measures breathing is a type that presses the chest with a band or the like, which causes inconvenience. In addition, the stress caused by such a sensor may affect the respiration measurement result. Methods: In order to solve this problem, there have been various studies such as amplifying and analyzing the fine movement of the image taken from a remote place, analyzing the temperature change of the thermographic image, and detecting the breath without attaching the sensor. However, these methods are very sensitive to motion and lighting, or use additional infrared cameras to detect region of interest in thermal imaging. In this study, we proposed a single sensor based method to measure the respiration rate of the user by analyzing the temperature change of the nose region by respiration after detecting the nose region through CNN (Convolutional Neural Network) in the thermal image as shown in figure 1. In order to detect the nose area, we used a model that has 18 layers based on ResNet and trained with 80000 pre-acquired thermal image data. This model receives an one-channel thermal image of 320×240 resolution as an input, and outputs the center coordinates and size of the bounding box representing the nose region, and whether the nose region exists. Figure1: The overall flow of the proposed method for single-sensor-based breath detection and its results. Results: The accuracy of the method for determining the presence of nose area in the image through CNN was about 99.8%. Actually, it was confirmed that the nose region where respiration rate analysis is possible exists in all images predicted as nose region. Time series breathing signals could be obtained by analyzing the temperature changes of the nostrils due to breathing and exhalation. The respiration rate estimated from the frequency analysis of the signal showed a correlation coefficient of 0.99 when compared to the ground-truth data obtained by the contact-type sensor. Conclusions: Our proposed method can be used for medical purposes to detect respiratory disturbances during sleep or to check the condition of psychiatric patients. In future studies, it is planned to improve the system so that it can be used more universally by measuring various other biological signals as well as breathing in the thermal image. Acknowledgements: This work was supported by the Industrial Strategic Technology Development Program (No. 10073159) funded by the Ministry of Trade, Industry & Energy (MI, Korea). Deep belief networks based stress classification model for stress monitoring using Korea national health and nutrition examination survey (KNHANES) VI Eunjeong Choi1, Dongkeun Kim2 1Department of Computer Science, Sangmyung University, Seoul, South Korea; 2Department of Intelligent Engineering Informatics for Human, Corresponding Author, Sangmyung University, Seoul, South Korea Objectives: Nowadays stress can be a fatal cause of various diseases and has become a big issue all over the world. An accurate prevention of stress is also meaningful in public healthcare and well-being life. In this study, we aim to design a stress classification model based on Deep Belief Network (DBN) method using public health dataset from the sixth Republic of Korea National Health and Nutrition Examination Survey. This model will be foundation of stress management that can be applied in daily life. Methods: Our proposed stress classification model consists of four steps. First step is data selection and preprocessing about sleep time, drinking, smoking, systolic blood pressure and body mass index data of people under 19years and 80years of age. Second, these preprocessed dataset undergo statistical analysis which consists t-test and chi-square test and clustering process. Third step is DBN modeling process. In this step, by setting the optimal hyper-parameter, the designed DBN based classification model is executed and the prediction of the stress level (low and high) is proceeded. Lastly, a performance evaluation are conducted. We compared the classification of stress to the other method; Naïve Bayesian (NB) and support vector machine (SVM). Results: The number of total input data is 1280 records. These input data was conducted t-test for continuous variables and chi-test for categorical variables; significance level was P-value 0.05. The variables at the significance level were gender, age, sleep time, pulse rate, height, weight, systolic blood pressure (SBP), drinking and smoking. The DBN model showed different results depending on the number of layers and the value of hyper-parameters. Table1 shows the accuracy and specificity values by the number of layers and hyper-parameter setting. Table1: The accuracy results in terms of different hyper-parameter settings in the designed DBN model The Naïve Bayesian and support vector machine methods showed the classification accuracy 65.23%, 66.02%, respectively. Conclusions: In this study, we classified stress using lifestyle and physical state data from the health questionnaire and nutrition survey data. Our DBN stress classification model\'s best accuracy is 69.67%; when set it to setting 2 of Table1. Compared to other methods, our proposed stress classification model based on DBN method is an effective and more accurate than NB and SVM methods. Therefore, we expect that the designed stress classification model is applicable for stress prediction in stress monitoring systems. Acknowledgements: This study was supported by a year-long school-funded grant (Sangmyung 2018). The computer-aided diagnostic system for visualization of drug-eluting stent on optical coherence tomography images Jong-Ha Lee Dept. of Biomedical Eng., Keimyung University, School of Medicine, Daegu, South Korea Objectives: OCT imaging tool allow us to understand more detail precise interaction between blood vessel and intra-coronary stents. However, OCT imaging only presented as single frame by frame. In this paper, we use image processing techniques on top of 3D OCT readings, to improve the visualization of DES invivo. Methods: The frames from the image file consisted of two parts: the top view panel showed the bisectional diameter of the stent and artery, where as the bottom view panel showed the lateral view of the stent and artery according to a pre-specified cross-sectional angle. The first task involved merely cropping into the top view panel, and removing the tick marks and labels. A circular transformation was conducted such that the center point of the artery was stretched along the top of the rectangular region, and the outer-most diameter was laid across the bottom of the rectangular region. This view set the edge of the artery on a straight horizontal orientation, and allowed an easier viewing of stent position, and arterial thickness. Results: Several visualizations were created a 3D reconstruction of arterial thickness, and a 2D contour plot with black pixel corresponding to location of the stent struts. Conclusions: The image processing and visualization techniques shown here enable easy identification of stent struts, and locations of thin arterial thickness. Acknowledgements: This research was supported by NRF-2017R1D1A1B04031182, Ministry of Trade, Industry & Energy (MOTIE, Korea) under Industrial Technology Innovation Program, No. 10063553, R0004840, R0006452, and KHIDI (HI17C2594). ¹Dept. Computer Science, Sangmyung University, Seoul, South Korea²Dept. Electronics, Sangmyung University, Seoul, South Korea Objectives: Gender recognition technology plays an important role as a component in a variety of fields including the healthcare field. Various data including speech and image data have been used in gender recognition. Especially, many gender recognition methods using human voice have been addressed. Existing gender recognition methods using voice data utilize uncompressed data. However, storing or transmitting human voice data always requires speech compression due to the significant amount. That is, the majority of speech data exists in a compressed form. Although gender recognition techniques using compressed speech data are highly needed, previously-known speech recognition techniques relied on uncompressed PCM data. Therefore, the objective of this paper is on gender recognition using compressed speech data. Methods: AMR speech compression is utilized to classify the gender in this paper because it is widely used in speech communications such as smart phones. The compressed data obtained by the AMR coding is constituted by frame units of a certain size. Bit information in a frame has different meanings depending on its position. By using this, the probability that each bit equals to one is used as a feature value, and feature vectors are formed by grouping the feature values for all bits in a frame. It has been observed in this paper that some portion of the feature vector has distinctly different forms for males and females. Results: First, five AMR data were used to obtain feature vectors for male and female. The feature vectors for male and female were obtained as shown in Fig 1. Second, a total of 20 AMR data for test were prepared, ten files are for male and another ten is for female. Here, the size of the AMR data is set to 1minute and the compression mode is set to 0. For the classification of gender, minimum distance comparison was performed where if the MSE between the feature vector and the input vector is less than 0.00634, it was classified to be male, otherwise female. Experimental results indicate that the detection performance for 20 male and female AMR data were obtained with 95% accuracy. Figure1: Feature vectors of 1-minute AMR data (a) for male (b) for female. Conclusions: Experimental results indicate that our method is able to recognize male and female voice in the compressed domain, not the waveform domain. Therefore, our method can be utilized in classifying the gender in-between communication channels or database systems in a fast fashion. 1Dept. of Computer Science, Graduate School, Sangmyung University, Seoul, South Korea; 2Dept. of Computer Science, Sangmyung University, Seoul, South Korea Background: Many diagnostic software collects information from facial expressions of patients, since the expressions reveal a lot of medical information about their status. The progress of ICT technologies builds various applications about remote medical examination and treatment. Therefore, an automatic system that recognizes facial components from a face on a photograph, such as eye, nose, mouth, chin and eyebrow, would improve the potentials of tele-medicine system in a great scale. Many deconvolutional structures based on supervised learning strategy have been proposed to present a semantic segmentation for a scene [1]. Methods: A deconvolutional network is a kind of neural network structure that attaches layers of deconvolution after the last convolutional layers [2]. Our structure is composed of four convolutional layers and four deconvolutional layers. At each convolutional layer, the resolution of the input tensor is reduced to 1/4, and the depth of the tensor is extended by 2. Therefore, an N x M x k dimensional tensor is converted to a tensor of (N/2) x (M/2) x (2*k) dimension. At each deconvolutional layer, the resolution of a tensor is extended to 4 and the depth is reduced by 1/2. Therefore, an N x M x k dimensional tensor is converted to a tensor of (2N) x (2M) x (k/2) dimension. We employ a hyperbolic tangent function for the activation function of our network. In each deconvolutional layer, the features extracted in the convolutional layer are attached to reconstruct the information of the tensor. Experiments: We have implemented our deconvolutional structure using Python and Pytorch library. For the training data of our network, we employ SCUT dataset [3]. We assign 70% of the SCUT dataset for training, 15% for verification and 15% for test. We apply mathematical morphology operations such as opening to remove unwanted artifacts from the results. Results: After 100 epochs of training, we show that more than 90% of the training dataset records 95% of IoU. Conclusions: We have designed a deconvolutional neural network structure for facial semantic segmentation. After we have applied a dataset of 55000 face images on our structure, we have shown that more than 90% of training dataset records 95% of IoU. From this result, our system can be applied in many tele-medicine systems based on face reading. Acknowledgement: This paper is supported by National Research Foundation of Korea (NRF) through NRF-2018R1D1A1A02050292 and NRF-2017R1D1A1B03034137. References: [1] H. Noh, S. Hong and B. Han, \"Learning Deconvolution Network for Semantic Segmentation”, Proc. of International Conference on Computer Vision 2015, pp. 1520-1528, 2015. [2] O. Ronneberger, P. Fischer and T. Brox, \"U-Net: Convolutional Networks for Biomedical Image Segmentation”, Proc. of Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2015, pp. 234-241, 2015. [3] L. Liang, L. Lin, L, Jin, D. Xie and M. Li, \"SCUT-FBP5500: A Diverse Benchmark Dataset for Multi-Paradigm Facial Beauty Prediction”, arXiv:1801.06345, 2018. A study on pillow design considered with materials, breathability and neck support based on document analysis Wei-Chung Chen1, Yann-Long Lee2, Chung-Shun Feng2, Syuan-Lan Shih2 1Guangdong-Taiwan College of Industrial Science & Technology, Dongguan University of Technology, Dongguan, China; 2Dept. of Industrial Design, Chaoyang University of Technology, Taichung, Taiwan Objectives: About one-third of a person\'s life is sleeping, if a person lives to 90years old, 30years are lying on a pillow, it can be seen how its importance. The Taiwan Academy of Physical Medicine and Rehabilitation (2006) pointed out that a good pillow can effectively support the cervical vertebrae, maintain the natural curve of the neck, reduce the pressure and tension of the joint muscles, relax the muscles of the neck and shoulders, and then get a comfortable sleep. Methods: The study uses Document Analysis, which is used for reading and organizing, describing, classifying, and interpreting to understand the past, insight, and predict the future of a pillow. For the earliest to modern pillows in China, considered the pillow height, material breathability, comfort, neck support, contact therapy, aromatherapy, etc for the design factors. Finally, using industrial design methods and processes, develop a health-care pillow that best meets for consumers. Results: The results found that since ancient times, people used to have the sleep habit with the pillow, it is said that primitive people use straw bales as pillows. The historian Sima Guang used small round wood as a pillow. When he slept a little while, he would slip off the pillow and immediately wake up, then he can continue to study hard. The name \"Warning pillow” is also a hard pillow. Pillows are also common in ancient medical books, Ming Dynasty \"The Compendium of Materia Medica” records: Bitter skin, black bean skin, mung bean skin, cassia seed, and chrysanthemum are used as pillows. Further, the pillow is given a medical function. In modern times, the development of pillow forms has become more diversified. Latex pillows, magnet pillows, memory foam pillows, etc., are born according to different functions. From the past to the present, pillows can be mainly divided into 3 categories, from hard pillows, medical pillows, and soft pillows. The results also found that the hard pillow is breathable and cleanable, but it\'s too hard to adjust and may cause damage to the neck. The general acceptance in the market is not high. Medical pillow added with herbs to enhance the additional function of the pillow, but there is still room for reinforcement in the support and comfort of the neck. Most of the soft pillows used by modern people only provide comfort functions. Most of it made of warm materials and have poor air permeability, which makes the head overheat and violation of the view that Chinese medical said: \"The head should be kept cool”. Some are still chemical materials, which are not beneficial to the human body. Conclusions: Indian health experts suggest that some techniques can be used to balance the five senses (auditory, tactile, visual, taste, smell) and can be used to improve sleep disorders. Finally, in view of the above advantages and disadvantages, researchers designed a health-care sleep pillow that conforms to the health science and can balance the five senses, make people healthier through sleep. Keywords: Pillow Design, Document Analysis, Health-care, Material of pillow. 1College of Information and Computer, Taiyuan University of Technology, Taiyuan, Shanxi Province; 2Virginia Polytechnic Institute and State University, United States Objectives: The difference in tumor size is significant and the segmentation accuracy is low for the two-dimensional (2D) convolutional neural network because it cannot extract the differences between the modal information for different image layers accurately. Single-mode Magnetic Resonance Imaging (MRI) cannot clearly reflect the different organizational structures of GlioBlastoma Multiforme (GBM). Therefore, this study aimed to propose a three-dimensional multi-pooled convolutional neural network based on the characteristics of brain tumors to solve the aforementioned practical problems. Methods: The convolutional neural network was applied to brain tumor segmentation. The multi-scale input and multi-scale sampling were used, and the conditions were used on the backend. The random field image segmentation was used as far as possible at the boundary to increase the accuracy of image segmentation and overcome the individual differences in brain tumors, while adapting to the size and position differences between different image layers of brain tumors. Results: The Dice coefficient was 91.64% through extracting information from multi-modal MRI images of 100 patients. Conclusions: The improved method of MRI brain tumor segmentation significantly improved the segmentation accuracy and could better extract the difference in information between each mode to achieve a wider range of MRI tumor segmentation. Trail of simplified EVH challenge for exercise-induced bronchoconstriction detection: a pilot study Lulu Wang1,2, Ahmed Al-Jumaily2 1Hefei University of Technology, School of Instrument Science and Opto-electronics Engineering, Department of Biomedical Engineering, Hefei 230009, China; 2Auckland University of Technology, Institute of Biomedical Technologies, Auckland 1142, New Zealand Background: The International Olympic Committee has recommended Eucapnic voluntary hyperpnoea (EVH) challenge with dry air as the standard method to identify exercise-induced bronchoconstriction (EIB) in elite athletes. However, existing EVH systems have some limitations including bulky and expensive implementation hardware components and not easy to use. To overcome these challenges, the authors recently developed a simplified method and breathing apparatus for the implementation of the EVH challenge. This pilot study aims to assess the sensitivity of the proposed simplified EVH method and breathing device to identify EIB in athletes. Methods: This human pilot study was approved by the ethics committee of the Auckland University of technology (Number 14/47). Twenty athletes were recruited from sports centers in New Zealand, including ten healthy subjects and ten subjects who previously diagnosed with EIB. Each subject performed the standard EVH challenge using Phillips apparatus and the proposed simplified EVH challenge using the developed breathing device on two different days. Airway responses were measured before and after each EVH test. Several forced expiratory volume in one second (FEV1) values were recorded before and in duplicate at 5, 10, and 15minutes after EVH test. The percentage changes in FEV1 were computed by taking the lowest FEV1 value after each EVH test. The subject was considered as EIB positive if FEV1 dropped 10% or more. Results: EIB was found in 11 subjects include ten subjects who previously diagnosed with EIB using the proposed method and breathing device. However, EIB was presented in 10 subjects using the standard EVH method and Phillips apparatus. Conclusions: The simplified EVH challenge method and breathing apparatus has the potential to become a useful clinical tool to identify EIB or respiratory diseases in athletes accurately. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grant No. 61701159), the Natural Science Foundation of Anhui Province (Grant No. 101413246, JZ2017AKZR0129), the Fundamental Research Funds for the Central Universities (JZ2018HGTB0236), the Foundation for Oversea Master Project from Ministry of Education, China (Grant No. 2160311028), and the funding from Hefei University of Technology (JZ2018HGTB0236). Objectives: With the deepening of scientific research, the comprehensive application of traditional Chinese medicine has attracted much attention, greatly improving the utilization rate of traditional Chinese medicine, turning waste into treasure and increasing social and economic benefits. This paper summarizes the comprehensive application value of jujube. Methods: This article reviews recent literature and data, the applications of jujube peel, jujube kernel, Jujube Flower and jujube in food, medicine and other aspects were reviewed. Results: Jujube is warm and sweet, it has the functions of Tonifying Qi and blood, strengthening spleen and stomach and so on. Jujube diet therapy also has the function of preventing and assisting the treatment of diseases. It can be used for tranquilizing the nerves and enriching blood. It is used as a Chinese medicine guide and a Chinese herbal formula for the treatment and prevention of diseases. In recent years, the food industry of jujube is also relatively hot, jujube slices, dried jujube, jujube drinks, jujube extract, jujube and wolfberry juice, jujube cake, jujube wine and so on are welcomed by consumers, and \"jujube seed” has become a pillar product of your jujube industry. The active components of jujube peel and jujube stone have been found and utilized gradually in the research work, but there are some problems such as low extraction rate. The extraction and application of edible pigment from jujube peel can avoid the hidden danger of carcinogenesis, poisoning and diarrhea caused by synthetic edible pigment. Adsorptive utilization of jujube stone provides a new way to purify industrial sewage. The discovery of antimicrobial effect of jujube stone wood vinegar opens up a new way for food preservation methods and types of preservatives. The production of jujube stone handicraft will enrich the kinds of natural jewelry, and reuse the waste jujube stone resources. The jujube stone bionics has also facilitated people\'s life and created social and economic benefits. At present, there are few studies on the pharmacological effects of chemical active substances from jujube peel and kernel, and the antioxidant mechanism is still unclear. Jujube stone and Jujube Peel Polyphenols can be used to reduce lipid, strengthen vascular wall, promote gastrointestinal digestion, and increase body resistance, anti-cancer and other effects. The jujube nectar in the Jujube Flower has a strong fragrance and high sweetness. It can moisten the intestines and lungs, tonify the spleen and stomach. It can also beautify the skin, detoxify and protect the liver. It is mainly used as dietary and health care products. Jujube bark, jujube stone, Jujube Flower and jujube tree contain similar active ingredients, but few of them have clinical application. Conclusions: The comprehensive utilization of jujube enlarges the scope of extracting active ingredients from traditional Chinese medicine, and the same active substances also exist in non-main medicinal parts. The comprehensive utilization of traditional Chinese medicine will promote the economic benefits of traditional Chinese medicine industry. The treatment of traumatic injury by Yang bud wine is simple, convenient and effective. It suggests that we can extend this kind of research to more traditional Chinese medicines, such as Honeysuckle, Achyranthes bidentata, Rhizoma Polygoni Corydalis and so on. To study the pharmacological action and clinical efficacy of its non-main medicinal parts is also conducive to the timely discovery of new medicinal and commercial value. Acknowledgements: This work was supported by the national major new drug discovery (2017ZX09301071), national international cooperation base (2016-15). Development of a plug-in for a fundus camera to simultaneously measure the oxygen saturation level and structures of the retinal vessels Peizheng Yan, Yongqing Wei, Yonghong Wang School of Instrument Science and Opto-Electronics Engineering, Hefei University of Technology, Hefei, Anhui, China Objectives: The retina is the only tissue in which vessels can be observed by the visible light. The color photo of retina captured by a common fundus camera only provides structural imaging. The relative retinal-vessel-oxygen-saturation provides metabolic imaging of the retina, which can be used to evaluate the pathophysiology of common eye diseases such as retinal vascular occlusions, glaucoma, diabetic retinopathy and age-related macular degeneration. This research designs a plug-in that expands the function of the common fundus camera. It only needs to take a retinal picture of the patient just as the common fundus camera does, but can obtain the oxygen saturation image and color photo of the retina at the same time, which makes it friendly to patients and doctors. Methods: On the basis of a common fundus camera, we remove the color camera and divide the optical path into two beams using a beam splitter. One beam is imaged onto a color camera for collecting the common color photo. The other beam is split into two beams again by another beam splitter, each of which passes through a narrow bandpass filter and is imaged onto a high-sensitivity camera. The center wavelengths of the bandpass filters are 570nm and 610nm respectively. Thus dual-wavelength images of the retina are obtained. In order to ensure the quality of the images, specially designed lenses need to be inserted before and after each beam splitter, which relay the original image to three images. Oxygenated hemoglobin and deoxyhemoglobin have different absorption rates for light at wavelengths of 570nm and 610nm, resulting in different vascular gray levels on the dual-wavelength images, then the oxygen saturation can be calculated by the difference. Results: A plug-in has been designed for the fundus camera from Chongqing Sunkingdom medical instrument Co., Ltd.. Twenty-two lenses has been used to relay the images and 45° field of view has been achieved. The optical resolution of central field of view, half field of view and edge field of view are greater than 60 lp/mm, 40 lp/mm and 25 lp/mm respectively. The image resolution of color photo and dual-wavelength images are 2304*1728 and 1936*1456 respectively. The oxygen saturation of vessels on the retina has been calculated by the dual-wavelength images. Conclusions: A plug-in has been designed using the relay lenses and spectroscopic optical system. It can provide the oxygen saturation and color photography of the retina simultaneously without increasing the operation difficulties, which will contribute to early diagnosis of retinal diseases. Acknowledgements: This work was supported by the Natural Science Foundation ofChina (No. 51805137), and the Natural Science Foundation of Anhui (No. 1808085QE129). Nuo Xu1, *, Yangcan Wang2, *, Junzhe Zhu1, Mengnan Ding3, Jie Liu3, Xuangjing Jin3, Yue He4, Lishan Shen5, Yinyin Jiang1, Yuxuan Wang6, Aloysius Wong7 1Department of Computer Science, Wenzhou-Kean University, Wenzhou, China; 2Department of English, Wenzhou-Kean University, Wenzhou, China; 3Department of Accounting, Wenzhou-Kean University, Wenzhou, China4Department of Mathematics, Wenzhou-Kean University, Wenzhou, China5Department of Economics, Wenzhou-Kean University, Wenzhou, China6Department of Psychology, Wenzhou-Kean University, Wenzhou, China7Department of Biology, Wenzhou-Kean University, Wenzhou, China; *Both authors contributed equally to this work Background: By 2050, drug resistance is projected to kill 10 million people, surpassing the amount of people killed by cancer annually. Recent reports from the World Health Organization revealed that microbes responsible for human diseases such as tuberculosis, pneumonia, malaria and salmonella, have become increasingly resistant to commonly prescribed antibiotics. Critically, these resistances have spread globally and are particularly rampant in developing countries due to the indiscriminate use of antibiotics in human-related activities such as food processing and agriculture. Currently, the lack of accessibility to localized as well as drug- and organism-specific information of antibiotic resistance has hampered efforts to combat the global spread of drug resistance. Here, we introduce Resistance Tracker (RT), which is a web-portal that tracks reports of antibiotic resistance across different organisms and sources. Methods: The RT web server was built using topic-specific knowledge bases and interactive networks generated through text mining from online repositories. PubMed-indexed articles were filtered through pre-determined keywords and matching articles were downloaded as MySQL database using the Medical Reference software. Duplicated articles were removed by a second screen with PubMed IDs. The RT database currently includes dictionaries of ‘Antibiotic Resistance’, ‘Bacteria’, ‘Animals’, ‘Human Body’ and ‘Locations’ built by surveying 70000 PubMed-indexed articles. Results: The RT web server is available at https://resistancetracker.com without registration or license. The tool allows users to search for the occurrence of resistance to a particular drug in one or more of the following fields: types of organisms which are further divided into bacteria, animals and humans, as well as geographical locations. The user first enters the type of resistance in the search box before selecting the desired fields. The web server then returns results containing resistance to the selected drugs in specified organisms or body parts if ‘human’ is selected as well as frequencies in selected geographical locations. Results are presented as ‘RT values’ which correspond with the number of articles that satisfy user-selected fields. Conclusion: In summary, we present the development of a new online tool that allows the tracking of resistance to 1) selected drugs in 2) specific organisms and at 3) selected geographical locations. This web server provides quick and convenient access to processed information of drug resistance thus enabling medical practitioners and the general public to make informed decisions. Given the increasing connectivity especially in developing countries, our web server contributes to the broader goal of combating antibiotic resistance. This tool is currently under development and will be revised based on user feedbacks. Please address all suggestions and comments to alwong@kean.edu. Acknowledgement: This work was supported by the Student partnering with Faculty (SpF) research programs of Wenzhou-Kean University (Grant No. WKU201617002 and WKU201718009). 1Hefei University of Technology, School of Instrument Science and Opto-electronics Engineering, Department of Biomedical Engineering, Hefei 230009, China; 2Department of Physiology and Biophysics, Mayo Clinic and Mayo Foundation, Rochester, MN 55905, USA Background: Microwave imaging is the relatively new imaging method, which has the potential to replace or become an additional method to mammography for early breast cancer detection. Over the past three decades, several microwave imaging techniques have been developed for the diagnosis of diseases including breast cancer, brain stroke, and lung cancer. However, existing microwave imaging techniques have some limitations including limited image resolution, long signal processing, and complex implementation instruments. Recent studies have shown that compressed sensing (CS) can improve biomedical image quality and reduce data collection time. The authors recently developed holographic microwave imaging (HMI) for identifying the dielectric object. This paper aims to develop a CS-based HMI method namely split Bregman-HMI (SB-HMI) for breast cancer detection in a fast and cost-effective manner. Methods: In this study, SB and orthogonal matching pursuit (OMP) algorithms are applied to produce a CS-based HMI image for identifying breast tumor. A computer simulation system, including a realistic breast model and imaging processing model, is developed under MATLAB environment to validate the SB-HMI theory for breast cancer detection. Various breast models are developed based on MRI images and published dielectric properties of tissues. Each breast model consists of skin, fat, gland, and arbitrarily shaped breast tumors. Phantom experimental validation is performed to evaluate the proposed SB-HMI by detecting small dielectric inclusions embedded in a tissue-like phantom. Both simulation and experimental results have been compared with various methods including HMI with the 16-element uniform sensor array, HMI with the 16-element random sensor array, SB-HMI with 10% measurements, SB-HMI with 30% measurements, and OMP-HMI. Results: Results show that SB-HMI with 30% or more measurements could fully recover the dielectric object and detect inclusions using much fewer measurements and less data collection time than other approaches. Some artifacts are displayed in the reconstructed image using SB-HMI with 10% measurements and OMP-HMI. Moreover, the proposed SB-HMI produces much fewer artifacts than other methods. Conclusions: The proposed SB-HMI approach could successfully identify small inclusions in a fast and cost-effective manner due to the implementation system requires much fewer antennas, which has the potential to become a useful computer tool for fast visualization and optimization of MI-based techniques. Acknowledgements: This work was supported by the Programme of Introducing Talents of Discipline to Universities of China (B12019), National Natural Science Foundation of China (Grant No. 61701159), the Natural Science Foundation of Anhui Province (Grant No. 101413246, JZ2017AKZR0129), the Fundamental Research Funds for the Central Universities (JZ2018HGTB0236), the Foundation for Oversea Master Project from Ministry of Education, China (Grant No. 2160311028), and the funding from Hefei University of Technology (JZ2018HGTB0236). Design of physiological parameter monitoring system based on unified Bluetooth protocol network Yongqing Wei, Feng Ji, Junxi Zhang School of Instrument Science and Opto-Electronics Engineering, Hefei University of Technology, Hefei 230009, China Objectives: Over the past two decades, several physiological parameter detection devices have been developed for monitoring health status in home and clinics. However, these devices have some limitations, including single-function, complicated, expensive, and unable to detect patients’ physical conditions according to actual needs. More importantly, previous developed physiological parameter detection devices are difficult for modern medical care applications. This paper aims to develop a new physiological parameter detection system using an unified Bluetooth to monitor health status accurately. It is hoped that the proposed system will become a powerful clinical tool for early diagnosis of disease and monitor health status. Methods: The authors applied Nordic Semiconductor\'s nRF series ultra-low-power SOC (System on Chip) to develop the detection module. The physiological parameter sensors were connected to SOC through the ADC, SPI, I2C and other peripherals on the chip. The physiological parameter information was processed, analyzed and stored using SOC in real-time. The measured parameters then transmitted to a central controller through a wireless Bluetooth-module embedded in the SOC. The central controller has aggregated all measured data into the shared database through the network. The detection module made of indicators and control buttons without screen display and UI. Such a design can minimize operation power, which makes the design more suitable for battery-powered occasions. The Bluetooth protocol contained a device model code, device status, packet type, and packet data. The maximum length of a single transmission packet was 256 bytes, and the maximum transmission rate was 115.2 Kbps, which satisfied the real-time medical device. The controller adopted various platforms such as Android, IOS, PC and embedded. Similar to the conventional devices, the proposed controller was connected to the detection module to achieve all UI functions and operations. This controller could work as a central node of the wireless Bluetooth network, and connect multiple required detection modules simultaneously to form a multi-parameter physiological parameter detection device to achieve more powerful functions. Moreover, the controller is a router that can aggregate data and share it with the network. Results: An nRF52840-based two-lead experimental ECG detector has been successfully developed. The detector can analyze and alarm nearly 20 kinds of arrhythmia, such as abnormal heart rate, premature beats and atrial fibrillation in real-time. The Android application could receive and display the detector\'s data just like an electrocardiograph. The comparison results are consistent with the Likang GWE10 dynamic electrocardiograph. A blood oxygen saturation monitor of the same platform was developed and gradually constructed into a complete physiological detection system. Conclusions: A physiological parameter detections system was developed using the Bluetooth protocol. Experimental results showed that the proposed system has the potential to replace existing physiological monitoring instruments and become an important carrier for telemedicine. Acknowledgments: This work was partially supported by the National Magnetic Confinement Fusion Science Program of China (Contract Nos. 2015GB103003) and the Doctoral Special Research Foundation of HFUT (Grant Nos. JZ2017HGBZ0942). 1Medical Electronics and Information Technology Engineering Research Center, Chongqing University of Posts and Telecommunications, Chongqing, China; 2Department of Dermatology, Chongqing Hospital of Traditional Chinese Medicine, Chongqing, China#The Author Contributed Equally to This Work and Should be Considered Co-first Authors Objectives: Brain stroke is a clinical emergency that needs rapid in-situ diagnosis and correct treatment in time. On the basis of self-built microwave detecting system, numerical analysis and experiments were conducted to discriminate ischemic from hemorrhagic strokes with high sensitivity. The proposed microwave system that can be carried in an ambulance as a prehospital detecting tool for brain stroke diagnosis. Methods: The operation of the system depends on the favorable dielectric contrast between brain tissue and blood that cause a dielectric change of head. The system employed a cylindrical waveguide sensor with wideband operation. And the detecting system was numerically analyzed to investigate the specific microwave-tissue interactions. Three dimensional printed anatomically realistic human head phantoms with actual dielectric tissue properties were fabricated for experimental validation of the proposed system. The phantoms had simulated hemorrhagic targets and ischemic targets inside with different size. Compared to the existing systems on the basis of classification technique and inverse scattering algorithm, the detection capability of the proposed system based on basic electromagnetic methodology was analyzed. The acquired time domain data was processed to differentiate ischemic from hemorrhagic stroke, as well as to measure the alteration of hemorrhagic/hemorrhagic stroke targets. Results: The numerical simulation results exploited the electromagnetic characteristics of head, and demonstrated the radiation safety of the proposed system. The experimental time domain performance of various head models showed that the differentiation of stroke could be diagnosed by measuring the change of group wave velocity inside the sensor. Moreover, the clinical efficiency of the proposed system was comprehensively elucidated with respect to, the variability of the dimensions and the dielectric properties of anatomical structures, and the positioning criticality of the system sensor around the head. Therefore, an effective diagnostic method for brain stroke had been realized. Conclusions: In short, this work showed a non-invasive system for diagnosis of acute stroke and verified it using a MRI derived 3D head phantom with respect to sensitivity and specificity, demonstrating the efficacy of the proposed detecting system in future preclinical trials. Acknowledgements: This work was supported by Scientific and Technological Research Program of Chongqing Municipal Education Commission (Grant No. KJ1704102); in part by Doctoral Foundation of Chongqing University of Posts and Telecommunications (Grant No. A2016-77). Experimental study on anti-adhesion mechanism of scalpel with microgrooved structure imitation shark skin Jing Xu1,2, Jiadi Lian2, Guodong Liu1, Jing Ni1 1Department of Mechanical Engineering, Hangzhou Dianzi University, Hangzhou, 310018, China; 2Hangzhou Steam turbine Company Limited, Hangzhou, 310014, China Background: The shield scale groove structure of the shark surface has significant drag reduction effect. The shark surface is covered by a diamond-shaped shield scale with a ribbed surface structure. The length of shield scales is usually 100˜200μm, and the width between ribs is 50˜100μm. The rib structure on the shield scale can optimize the fluid state of the fluid boundary layer on the shark surface, suppress and delay the occurrence of turbulence, and effectively reduce the water resistance. At present, the special structure of shark skin surface is widely used in aerospace and ship antifouling industry, but it is rarely used in medical field. In this study, the anti-adhesion mechanism of scalpel with micro-groove structure of shark skin is studied. Methods: Firstly, scanning electron microscope and surface profilometer are used to observe and measure the microstructure of shark skin, and combining with the basic hydrophobic theory, a modified Cassie model is established to analyze the hydrophobic mechanism of the groove structure theoretically. The contact angle of the shark skin surface and the shark surface was measured by using the micro-electroforming and high speed camera. The anti-stickiness of scalpel with micro-grooves was studied from two aspects: theoretical analysis and experimental measurement. Results: According to the Cassie modified model, it is found that if the ridge height of micro-groove rib on the shark skin surface is smaller, and the length and width of scale are smaller, the static contact angle is larger and hydrophobicity is stronger. The interface contact between liquid drop and shark skin surface is shark skin liquid drop and gas liquid drop compound contact angle, thus increasing the contact angle of liquid drop on shark skin surface and increasing hydrophobicity of shark skin surface. The shark skin surface obtain the anti-adhesion effect. Conclusions: Shark skin microgroove scalpel can effectively prevent the scalpel from sticking to human tissues and blood during surgery. Moreover, as the ridge height of the micro-groove ribs on the surface of the shark-like scalpel is smaller, the anti-sticking effect is better. Acknowledgements: The research is financially supported by the National Nature Science Foundation of China (No. 51505112, 51775153), the China Postdoctoral Science Foundation (No. 2018M632436) and the Zhejiang Province Public Welfare Technology Application Research Project (No. LGF18E060001). The prediction and diagnostic value study of CT for lymph node metastasis, distant metastasis and local lung cancer invasiveness Yingmin Chen1, Yunfeng Bao1, Hongming Zheng2, Xiaodan Zhang3, Congcong Hu1 1Hebei General Hospital, Hebei, China; 2The Second Hospital of Hebei Medical University, Hebei, China; 3Science and Pharmacy Department of Hebei Thoracic Hospital, Hebei, China Objectives: To explore the value of CT in predicting lymph node metastasis, distant metastasis and local lung cancer invasion, and to analyze its role in clinical staging of lung cancer. Methods: From January 2017 to January 2018, collecting 158 patients with non-small cell lung cancer undergoing lymph node biopsy underwent radical resection in our hospital. All patients underwent CT examination before operation, and were diagnosed and staged according to the results of operation. We calculate the positive predictive value and diagnostic value of CT in the diagnosis of lymph node metastasis, distant metastasis and local invasion of lung cancer. Results: there is that 937 lymph nodes (N1, N2 and N3) were confirmed by mediastinoscopy in 158 patients. The sensitivity, specificity, positive and negative predictive values of CT in the diagnosis of lung cancer lymph node metastasis are 51.0%, 76.1%, 49.0%, 77.6%, and The accuracy is 68.4%;The sensitivity, specificity, positive and negative predictive values of CT in the diagnosis of distant metastasis are 83.7%, 89.0%, 77.4%, 92.4%, and The accuracy is 68.4%;The sensitivity, specificity, positive and negative predictive values of CT in the diagnosis of local invasion are 85.3%, 92.1%, 82.4% and 93.2%, and The accuracy is 92.4%. Conclusions: CT has high accuracy in predicting lymph node metastasis, distant metastasis and local invasion of lung cancer and it has important diagnostic value. The clinical effect of C7 nerve root transposition in the anterior approach for repair of brachial plexus injury Bo Chen, Shoucheng Wu, Shuai Jiang The First Affiliated Hospital of Zhejiang Medical University, Zhejiang, China Objectives: To observe the clinical effect of C7 nerve root transposition in the anterior approach for repairing brachial plexus injury. Methods: Choosing 21 operations were performed in our hospital from March 2002 to August 2003. 14 cases were followed up for 1year. Among the 14 cases, there were 12 males and 2 females, the aged is 17-41years, and the average age is 31years. Operative time is 2-6months after injury; the average time is 4months. There were 5 cases is the brachial plexus avulsion, 8 cases is upper and middle trunk avulsion accompanied by incomplete injury of the lower trunk, and 1 cases is upper and middle trunk sharp injuries. The C7 nerve roots of the healthy side were cut off at the junction of the trunk and femur, and the proximal end was free to the intervertebral foramen. The C7 nerve roots were reversed to the anterior esophageal posterior space of the vertebral body through the anterior trapezius muscle and bridged with 4-6 femoral cutaneous nerves to repair the upper trunk of the affected side. Results: 14 patients were followed up for 12-19months, the average is 16months. 10 cases is The healthy side C7 nerve root repair the upper side of the trunk or C5, C6, 9 cases had shoulder abduction and flexor elbow muscle strength grade 3, 1 case had no effect; 3 cases had upper trunk posterior thighs repaired by contralateral C5, 3 cases had shoulder abduction muscle strength grade 3, and 9 cases had contralateral C7 and nerve root repaired C5. Biceps brachii muscle strength is grade 0. Conclusions: The contralateral C7 nerve root can be transferred through anterior vertebral approach to repair the superior trunk of brachial plexus and to reconstruct shoulder abduction and elbow flexion by repairing the posterior femur, and obtain good results. The application study of spiral CT enhanced scanning on venous trocar and high pressure syringe Yunfeng Bao1, Hongming Zheng2, Xiaodan Zhang1, Yingmin Chen1, Cunhui Shan1, Xiaoying Huang1 1Hebei General Hospital, Hebei, China; 2The Second Hospital of Hebei Medical University, Hebei, China Objectives: To explore the application of intravenous trocar and high pressure syringe in spiral CT enhanced scanning. Methods: Choosing 214 cases of CT enhanced scan with venous trocar puncture, and observing the leakage of contrast medium .According to the random number table method, it was divided into observation group and control group, 107 cases in each group. The patients in the observation group were treated with disposable safe intravenous indwelling needle, while those in the control group were treated with disposable butterfly pterygoid scalp intravenous needle. The incidence of needle injury in nurses during puncture was compared between the two groups, and the pain score, the number of successful puncture cases, the success rate of scanning and the occurrence of leakage were compared between the two groups. Results: In the observation group, the nurses did not suffer from needle injury during puncture, while in the control group, 11 cases (5.5%) suffered from needle injury during puncture, with a significant difference (x2=1.95, P=0.034). The pain score at the puncture site in the observation group was (3.1±0.52), while that in the control group was (6.3±0.85), with statistical significance (P 0.05). The success rate of one puncture in the observation group was 95.0%, which was higher than that in the control group 90.5% (P 0.05). In the observation group, there were 3 cases of contrast medium extravasation during the high pressure injection. In contrast, 16 patients in the control group had a small amount of extravasation of contrast medium during high pressure injection, and 3 patients had a large amount of extravasation of contrast medium. The difference was statistically significant. Conclusions: Safe intravenous indwelling needle can improve the success rate of puncture, reduce the leakage rate of contrast medium, protect the occupational safety of nurses, and reduce the incidence of occupational acquired infection caused by needle injury. At the same time, it is convenient for patients to exercise their limbs, and is easy to rescue and give medicine when complications occur. Additional insight into knee joint protection: The influence of knee-pads on lower limb muscles Zhi Tang1, Chang Xu2, Zhihui Liu2, Qian Wang3 1Haipai fashion design and value creation collaborative innovation center, Donghua University, Shanghai, China; 2College of Mechanical Engineering, Donghua University, Shanghai, China; 3Shanghai First People\'s Hospital, Shanghai, China Objectives: The counterforce from the ground when jogging will reach the triple of our body weight and our knee supports 70% of it. Wearing a knee-pad is an effective resolution to avoid secondary damage. The pathogenesis of knee arthritis is closely related to the abnormal biomechanics of muscles. Muscle plays an important role not only in joint movement but also in joint protection. The purpose of this study was to research the effects of different types of knee-pads on fatigue of lower limb muscles when people are jogging. Methods: Ten healthy college students (5 female and 5 male) volunteer to participant in experiment with the age of 21±5years and a height of 165.5±8cm. The volunteers jogged in the treadmill with Closed Patellar Knee Support (CPKS) and Patellar Tendon Strap (PTS) respectively. Tibialis anterior muscle, musculus gastrocnemius, rectus femoris and musculus vastus lateralis were chosen as the lower limb muscles for our study. The surface electromyographic signal (sEMG) about four lower limb muscles were collected by TMSI Bluetooth wireless multi-parameter biofeedback instrument. Results: Time domain analysis and Frequency domain analysis were used to analysis the data result. The single factor analysis of variance is used to compare differences between two kinds of knee-pads. The RMS (root mean square) was calculated according to time domain analysis. The activity conditions when wearing with CPKS are lower than PTS overall (P 0.05). The restraint of CPKS on knee joints is rather tighter. Through comparing the highest RMS of rectus femoris and musculus vastus lateralis. It can be still obviously observed that the muscles’ activity level of wearing CPKS is lower than that wearing PTS. Through the Frequency domain analysis, the mean power frequency (MPF) of gastrocnemius muscle and rectus femoris decreased by 15.8% and 11.2% respectively (P 0.05). The MPF of vastus lateralis muscle and anterior tibial muscle decreased by 13%, 12.8% (P=0.027 0.05, P=0.048 0.05). In the condition of wearing the PTS, the MPF of anterior tibial muscle, gastrocnemius, rectus femoris, vastus lateralis were decreased by 5.8%, 3.6%, 4.8%, 8.5% (P 0.05). It can be integrated that the muscles would be more fatigued when wearing CPKS. Conclusions: It is obviously that the lower limb muscles be more fatigued if the knees were worn with Closed Patellar Knee Support (CPKS). It is more suitable for the people who suffering from rheumatism. Then, the Patellar Tendon Strap (PTS) can play a protective role in the patella, with no direct protective function on other parts of the knee. So, the PTS may be unsuitable injury disease that in lower limb muscles. Acknowledgements: This project is funded by Shanghai Municipal Science and Technology Committee (Grant No. 18dz2301400). Fundamental Research Funds for the Central Universities. Objectives: Medical images can reveal the cause and the severity of disease, which have important applications in clinical diagnosis. The ambiguous corresponding relationship between the local area and pathological features brings difficulties for establishing the computer aided diagnosis model. Based on the global image feature expression and uniform method, it is difficult to effectively express the local pathological features. There is a big difference between the feature description of the bottom layer and the semantic description of the top layer, which makes the performance of the existing image annotation algorithms not very ideal. To make effective use of these heterogeneous features in the process of image annotation, an automatic annotation method of medical image semantics based on deep learning is proposed. Methods: The proposed method can free people from laborious manual annotation; and it overcomes the gap between visual features and semantic expression by establishing the mapping between low-level visual features and high-level semantic, so as to realize the retrieval application of massive medical image data based on semantics. It uses the mapping model of annotated images automatically learning semantic concept space and visual feature space, to establishes the mapping relationship between low-level image visual feature and high-level semantic feature. And the visual feature network parameters are trained based on the deep learning. The purpose is to extract highly reliable visual features that can accurately describe the image content. Then parameter optimization and semantic mapping model training are carried out based on the manually annotated training set. It includes three steps: extracting the corresponding visual features of training set images by using the network parameters of visual features obtained by training; Semantic label classifier is obtained based on visual feature training; By taking the semantics of the artificially annotated training set as the labeling benchmark. The objective function is established through the semantic comparison with the semantic label classifier based on the convolution neural network method. Where, the model constructed by the network parameters of visual features and the parameters of semantic label classifier is the image model between low-level visual features and high-level semantic features. Results: In order to verify the universality of our method, we test on the medical image from Dalian Medical University. The experiments are conducted in 15 categories with our proposed method. The annotation accuracy is approximately 89.2%. The results show that the proposed method is a effectiveness method based on deep representation learning. Conclusions: In this paper, the medical image semantic description model and automatic semantic annotation framework based on deep learning can effectively extract the visual features of image content, set up low-level visual features and high-level semantic mapping and overcome the gap between the visual features and semantic expression, which realizes the semantic retrieval in huge amounts of medical imaging. The experimental results show that the method is superior to the state-of-the-art image annotation methods. Classification of Mongolian medicine Polygonatum Sibiricum metabolomics data based on neural network Xiye Wang1, Mingyang Jiang2, Zhili Pei2 1Chemistry and Chemical Engineering College, Inner Mongolia University for the Nationalities, Tongliao, China; 2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: The object of this study is metabonomics data on the mechanism of Mongolian medicine Polygonatum Sibiricum. In this paper, a metabonomics data classification method based on neural network is proposed. This method can effectively accomplish the judgement of disease and the curative effect of Mongolian medicine Polygonatum Sibiricum. Methods: There were 26 healthy Wistar male rats, each weighing approximately 200g. After 1week of adaptation to the environment, they were randomly divided into group C (normal group) (n=13) and group P (yellow group) (n=13). After the process of intragastric administration, blood collection and centrifugation, the spectral data was completed by ultra performance liquid chromatography-mass spectrometry. Chromatographic conditions: the temperature of the column is 40°C, and the binary linear gradient elution contains 0. 1% aqueous solution of formic acid, methanol. Mobile phase gradient composition: 0˜2min, 8%˜80% mobile phase B; 2˜5min, 80%˜100% mobile phase B; 5˜7min, 100% mobile phase B; 7˜8min, 100%˜8% Mobile phase B; 8˜10min, 8% mobile phase B. The flow rate was 0. 4ml / min, and the sample injection volume is 10μl. Mass spectrometry conditions: ESI ion source, positive and negative ion mode, scanning range: m/z 100˜1000, positive ion mode: capillary voltage: 3kV, cone voltage: 40V, ion source temperature: 100°C, atomization temperature: 400°C, shell gas flow rate: 800L/h; negative ion mode: Capillary voltage: 2. 5kV, cone voltage: 40V, ion source temperature: 80°C, atomization temperature: 150°C, shell gas flow rate: 600L/h. Results: Ultra-high performance liquid chromatography-mass spectrometry data was used for peak extraction, peak alignment and normalization based on Masslynx V4.1 software. Biological databases such as HMDB and METLIN were used to perform biomarker identification and analysis of metabolic pathways. After obtaining the normalized data, the neural network model was used to complete the dimensionality reduction, the activation function was ReLU, the learning rate was 0.001, and the classification was completed by the Softmax classifier. The experiment used a 5-fold cross-validation method to complete the evaluation of the classification model. Conclusions: Compared with other classification method, Neural Network can effectively complete the classification of Mongolian Medicine Polygonatum Sibiricum metabolomics data. The experimental results show that the classification results of the proposed method are better. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). Optimization of ultrasonic-assisted extraction of Russia-polysaccharides based on Russia Markov chain Jie Gao1, Lin Huang2, Jianjun Xu2 1College of Foreign Languages, Northeast Petroleum University, Daqing, China; 2College of Electrical Information Engineering, Northeast Petroleum University, Daqing, China Background: By Use 5year data, Trichosanthin (TCS) is a Chinese herbal medicinal that has shown hepatoprotective effects with anti-inflammatory, anti-oxidants, and antiviral activity. A new Fuzzy Markov biologic chain algorithm is given for an incomplete measurable discrete system, for example, Trichosanthin. The objective of this study is to optimized the ultrasonic-assisted extraction of polysaccharides from TCS and then studied the antioxidant capacity of the extracted polysaccharides. Methods: According to the period for RUSSIA FUZZY Markov Chain system, on the basis of single factor experiments, the Box-Behnken experiment analyzed and optimized the ultrasonic-assisted extraction of polysaccharides from TCS by response surface methodology. The antioxidant activity of the extract\'s polysaccharides was then determined including the DPPH free radicals and scavenging activities against hydroxyl (˙OH), superoxide anion (˙O2_) and total antioxidant reactivity. The main purpose of preprocessing the hydroxyl (˙OH), superoxide anion (˙O2_) image is to highlight only the area of interest and debilitate areas that are not interested. The patrol image preprocessing includes image gray-scale processing, optical correction, image restoration, image noise reduction, and the like. The function of the rgb2gray in the bioDBnet toolbox can accomplish this type of operation. Results: The optimum conditions upon ultrasonic extraction were as follows: ultrasonic temperature 45°C, ultrasonic time 55min and water-to-material ratio 15mL/g. It will establish a combinational evaluation model based on fuzzy Markov network of Russia medical science pattern changes. it can automatically adjust the detection range to track the object and achieve the function of automatically acquiring images. Conclusions: The polysaccharides could be as a good antioxidant agent in different foods, cosmetic and pharmaceutical formulations. The grayscale image converted from the color image retains most of the image information, for example, TCS, which reduces the amount of computation in subsequent image processing and effectively improves work efficiency. Acknowledgements: This work is Supported by Northeast Petroleum University Innovation Foundation for Postgraduate (JYCX_CX06_2018), and the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Jianjun Xu. A device for blood pressure estimation from oscillatory pressure pulse wave and PPG signal based on the neural network Jian Tian1, Jiangan Xie1, Zhonghua He1, Yuhao Jiang1, Yi Lu2, Xiuxin Wang1 1Medical Electronics and Information Technology Engineering Research Center, Chongqing University of Posts and Telecommunications, Chongqing, China; 2School of Optoelectronic Engineering, Chongqing University of Posts and Telecommunications, Chongqing, China Objectives: The accurate blood pressure (BP) measurement contributes to the prevention of cardiovascular disease, a leading cause of death globally. Wrist cuff BP monitors are widely accepted by the people with health needs. However, the accuracy of these devices has been questioned. Their measurement results are easily influenced by the wrist movement disturbance. By contrast, the finger photoplethysmography (PPG), which provides the hydrodynamics information of arterial system, is less affected by wrist motions. In present study, the features of the PPG Signal and oscillatory pressure pulse were extracted and fed into a genetic algorithm-back propagation (GA-BP) neural network for the BP estimation improvement. Methods: A wearable BP monitor integrated with a PPG sensor was fabricated. In the measurement procedure, when the wrist cuff pressure decreased linearly with time, the oscillatory pressure pulse wave and the finger PPG signal were recorded simultaneously. The peaks and valleys of the two signals were detected and the amplitudes of the oscillatory pulses were normalized by dividing by the maximum value. Then, the normalized pressure pulse amplitudes were fitted to Gaussian curves, while the PPG pulse amplitudes were fitted to a polynomial curve. Parameters of the curves and the root mean square errors of the curve fitting were used as the input of a GA-BP neural network. The time interval between the pressure pulse peak and the corresponding PPG pulse peak, relative to the pulse wave velocity, was selected as an input of the neural network. The training data set (30 subjects) and testing data set (20 subjects) were acquired from healthy users with the systolic and diastolic BP ranges of 89 to 127mmHg and 60 to 85mmHg, respectively. The subjects were measured with the developed device and subsequently with the mercury sphygmomanometer. To compare the performance of the developed device and the commercially available wrist BP monitor, the accuracy of the Omron HEM-6111 BP monitor was tested. The differences between the wrist devices and mercury sphygmomanometer were investigated by Bland-Altman plots. Results: The differences between the device and the mercury sphygmomanometer were 0.7±2.6mmHg (mean±standard deviation) for systolic and -1.5±3.1mmHg for diastolic BP, and the differences between the Omron HEM-6111 and the mercury sphygmomanometer were -2.9±3.6mmHg for systolic and 3.23±4.42mmHg for diastolic BP. Conclusions: The developed device, insensitive to the wrist motion artifacts, showed an improved performance. However, a mathematic model for the finger PPG signal during cuff deflation should be built for theoretical supports. Acknowledgements: This work was supported by the PhD Start-up Fund of Chongqing University of Posts and Telecommunications (Grants No. A2017-133). Chongqing Key Laboratory of Photoelectronic Information Sensing and Transmitting Technology, Chongqing University of Posts and Telecommunications, Chongqing, China Objectives: Blood glucose level of the body is an important physiological parameter to measure, but is challenging to measure it accurately by needle-free approach. One of the technology to offer non-invasive, continuous glucose monitoring is reverse iontophoresis, which can extract interstitial fluid though the skin via electroosmosis. In recent years, Wearable sensing technologies have received considerable interests due to the promising use for real-time monitoring of health conditions. MXene, a family of 2D transition metal carbides and nitrides, are preferred material because of their high electrical conductivity and flexibility. Here, we report a wearable glucose sensor based on a flexible MXene (Ti3C2Tx) pixel array electrode platform. Methods: In this work, etching-induced exfoliation approach is used to prepare different layers of MXene. To form a stretchable electrode, the MXene are casted in an interdigitated pattern with finger length of 10mm and spacing of 2mm and embedded below the surface of PDMS. The glucose is extracted from the interstitial fluid via electroosmotic through individual, follicular pathways in the skin. The glucose sensor is calibrated against a commercial glucose level system on an artificial skin over a wide glucose range. The system is packaged into a flexible wristband, and the circuitry on the wristband is composed of a network analyzer chip, a button cell battery, and an ultralow power Bluetooth module wirelessly paired with a smartphone. The data can also exportable to researchers and medical professionals for in-depth analysis and remote health monitoring. In addition, some positions are reserved for other sensors, such as skin hydration sensor and electrocardiography electrodes. Results: The electrodes can conformal contact to the skin. This system can efficiently detect the physiological signals of glucose in human body and the data can be collected successfully by a mobile phone. The device is highly flexible and maintains structural integrity and detection ability under repeated deformations, including bending and twisting. Conclusions: This report proposes a unique system for convenient wearable healthcare monitoring, which is useful in clinic and may be commercialized for diversified applications in the future. Acknowledgements: This work was supported by Scientific and Technological Research Program of Chongqing Municipal Education Commission (Grant No. KJ1704102); Doctoral Foundation of Chongqing University of Posts and Telecommunications (Grant No. A2016-77, A2018-10, A2018-21). High strength translucent lithium disilicate glass-ceramics for dental implant application Wenyan Luo, Qiankun Xiao, Xiaofeng Dong, Zhuohao Xiao School of Materials Science and Engineering, Jingdezhen Ceramic Institute, Jiangxi, China Objectives: Lithium disilicate glass-ceramics are widely used in dental restoration and transplantation because of its tooth-like luster, adjustable color, and good biocompatibility with human body. However, the inherent brittleness of glass-ceramics is the bottleneck for dental application. Improving the strength and hardness of lithium disilicate glass-ceramics is the key to ensure the service life of dentures. In order to improve user experience and prolong service life of the denture, we propose an ion-exchange treatment to further improve the strength of the glass-ceramics. Methods: Li2O-SiO2 matrix glass was prepared by using the conventional melt-quenching method, with analytical reagent of Li2CO3 and SiO2. The mixed raw material was melted in an alumina crucible at 1400 °C for 2h in air atmosphere. The homogeneous melts were poured onto a preheated stainless steel mold and annealed at 400 °C for 2h, followed by cooling down naturally to room temperature, then transparent and colorless Li2O-SiO2 matrix glass was obtained. The matrix glass was cut and polished into desired shapes, followed by nucleating at 550°C for 2h and crystallizing at 700°C for 1.5h to synthesize translucent lithium disilicate glass-ceramic samples. The samples were then immerged in KNO3 molten salt at 450°C for different time. The large radius K+ in molten salt is exchanged with the small radius Li+ in the glass phase of lithium disilicate glass-ceramic samples, which forms a layer of compressive stress on the surface of the samples and improves the strength and hardness of the surface. Results: The bending strength of the matrix glass after casting is only 110MPa. By controlled crystallization treating,the matrix glass transformed into glass-ceramic and its bending strength obviously increased to 265±10MPa. SEM images show clear interlocking Li2SiO3 crystal structures in the samples. The high temperature ion-exchange with KNO3 molten salt can further enhance the bending strength of lithium disilicate glass-ceramics. When the exchange time is 6h, the bending strength of lithium disilicate glass-ceramics can reach 412±10MPa. However, the Vickers hardness of treated glass-ceramic samples is only 6.5±0.3 GPa, which has not been significantly improved. Conclusions: By ion-exchange treatment of lithium disilicate glass-ceramics in KNO3 molten salt, the Vickers hardness of glass-ceramics has not been effectively improved, but the bending strength of the material can be increased by more than 50 %. The glass-ceramic with the highest strength of 412±10MPa could be obtained when the exchange time was 6h. The ion-exchange process of lithium disilicate glass-ceramics is expected to be widely used in dental implants applications. Acknowledgements: This work was supported by the National Natural Science Foundation of China (51762023), the Jiangxi Provincial Department of Education (GJJ170801), and the Training Program of Outstanding Young Scientists in Jiangxi Province (20171BCB23070). Research on the strategy of soc-based cancer by Raman spectroscopy and principal component analysis Hongyu Li, Chao She, Shuang Ren, Jianjun Xu, Shanfeng Ding, Yue Liu, Limei Yan School of Electrical Engineering and Information, Northeast Petroleum University, Daqing, China Background: With the development of Raman spectroscopy, it has become one of the main methods to treat esophageal cancer because of unhealthy diet. it has been more and more widely used in biomedicine, biophysics, chemistry, pharmacy, and so on. Methods: The Fourier-Transform Raman spectroscopy is used to diagnose and monitor the hemoglobin in SOC-based. Principal component analysis (PCA) is employed to discriminate the healthy and cancer blood samples. Raman spectroscopy was acquired from 35 individuals with esophageal cancer and healthy patients. It can be seen that the maximum frequency deviations of the system without accessing to EV and the system with accessing to EV are about 0.7Hz and 0.5Hz, respectively. By comparison, the maximum frequency deviation with accessing to EV is reduced by 29%. Regarding the system without accessing to the EV system, the restoration time for the reference frequency attachment is 8seconds under the action of the frequency modulation unit. In contrast, the restoration time of the system is only 4seconds under the same disturbance upon accessing to EV, which is reduced by 50% of the overshoot compared to that without assessing to EV. It\'s observed that the system frequency can be quickly restored to the safe range when frequency is disturbed after the accessed participating in system frequency modulation. Thus, EV participating in system frequency modulation has a actual effect. Results: Based on calculating examples of the Matlab/Simulink simulation, the control strategy proposed could effectively suppress the EV cancer. The results show that the intense of the peaks at 1562cm_1 in the spectra of healthy hemoglobin are higher than the esophageal cancer hemoglobin, while it is just the opposite in the spectra of esophageal cancer hemoglobin. The diagnostic accuracy of Raman spectroscopy is about 90-95%. Conclusion: The simulation results have shown that using EV can effectively suppress fluctuations of cancer and shorten the time frequency modulation. Acknowledgements: This work was supported by the project of Post Subsidy Funds for the Sci & Tech Achievements of Northeast Petroleum University (HBZZJ201603), and the National Natural Science Foundation of China (51774088), and the Innovation Fund of China National Petroleum Corporation (2016D-5007-0201). The Corresponding Author is Shuang Ren and Jianjun Xu. 1Key Lab. of Enhanced Oil and Gas Recovery of The Ministry of Education, Northeast Petroleum University, Daqing, China; 2College of Earth Sciences, Northeast Petroleum University, Daqing, China; 3Zhanjiang Branch Company of CNOOC Company Limited, China Background: Currently, Body attributes analysis technique can resolve data into attribute information, transform the information to lithology, physical property and fat gas property, which is the effective means for medical prediction. Because there is a lot of vegetation in the inspection route, there are a lot of interference factors in the Medical Prediction image, and the noise and vibration blur are very easy to appear in the images collected by the Medical Prediction, which causes the deterioration of the image quality. And is influenced by human conditions, lighting conditions during image acquisition inspection cannot be guaranteed, so that uneven brightness of the image captured image happens, it is necessary to preprocess the image acquisition. Methods: In Block Xing10, the log of calcium medical layer shows \"three-high” characteristic, which refer to high resistivity, high density and high velocity(low time difference). According to this feature, the calcium blood pressure image, velocity of blood fat and gall stone, and density of wells are calculated in PI21b of sedimentation unit and adjacent layers. Then four medical models based on those parameters are built. Using mathematical theory optimized along layer amplitude and amplitude difference to Medical predict reservoir, propose the method of attribute threshold value automatic determination with seismic attribute and well-to-seismic integration inversion. According to the attribute slice obtained from threshold value can reflect the body distribution, and the rate of people attribute results in study body area unit, the prediction accuracy of body reach to 85%. Results: (1) The threshold value quantitative representation method based on body attribute is improve work efficiency and accuracy. (2) The optimum results of every body unit are along layer amplitude and amplitude difference, which can predict formation better. Conclusion: As a result of the action of biofilm, establish the stratigraphic modeling with calcium layer, seismic forward results show that calcium layer can influence the coincidence rate of reservoir prediction. Acknowledgements: This paper is financially supported by the National Natural Science Foundation of China (No. 51804078) and the University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province (UNPYSCT-2017031). The Unattended-valued (UAV) image acquisition technology using in immediate power medical examination Aihua Xu1, Yang Du1, Yankai Xu2, Limei Yan1, Lin Huang1 1College of Electrical Information Engineering, Northeast Petroleum University, Daqing, P.R. China; 2Ruixin College, Beijing Institute of Technology, Beijing, China Objectives: The vibrations and other factors that occur due to the relative motion between the inspection target and the image acquisition device and the influence of the drone itself and other factors may cause medical examination image distortion. So, Image processing technology is one of the most important technologies in Unattended-valued (UAV), immediate power medical examination (IPME) can greatly improve the quality of image recognition and improve the accuracy of fault diagnosis. Methods: The medical image target with different shapes and different fault types were selected as the training samples of the model. There were 1000 pictures of each type of medical image target, 1000 pictures of different fault types, and a total of 10000 training samples. In the experiment, the complete medical image target appearing in each picture in the training set will be marked, its peripheral frame will be marked, and its position will be recorded, giving a category label. For the test samples of each type selected 500 images as a test set and the training sample must not appear in the test set. Because the images collected by the Unattended-valued (UAV) medical examination are mostly color images, and the color of the inspection image itself is also a kind of useless information for the subsequent processing. Therefore, the process of converting a color image to a gray image by a specific algorithm is easy. The function of the rgb2gray in the MATLAB toolbox can accomplish this type of operation. Results: The correctness of this method for various types of immediate medical examination and different types of faults can reach more than 95%, the recall rate is more than 78%, and the missing rate is below 15%. When the training samples are increased to 20000. After training the model, the correct rate can reach 98.8%, the recall rate is 88.2% and the missing rate is 11.5%. Conclusions: This method can effectively improve the efficiency of inspection of immediate medical examination, making the examination more intelligent. it can also play a guiding role in other related fields. With the continuous development of deep learning, the intelligence and efficiency of drone inspection will gradually become a reality, I believe in the near future, these areas will become challenging and the focus of attention. Acknowledgements: This work is Supported by Northeast Petroleum University Innovation Foundation for Postgraduate (JYCX_CX06_2018), and the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Limei Yan. College of Electricity Information Engineering, Northeast Petroleum University, Daqing, China Objectives: At present, the pore diameter distribution is an important part of the characterization of Cells pore structure. In this paper, the pore diameter distribution characteristics of Neuromedicine are described in a comprehensive and intuitive way through the method of image observation. Methods: By used the laboratory experiments and the theoretical models, Neuromedicine in Longmaxi formation of southern Sichuan was selected as the research object. Then, pore area density function was acquired by fitting the data. The conversion model between area density function and volume density function was established and the pore volume density function was obtained. And by changing the existing images in the original training set to different angles, the number of training samples is increased to 20000. After training the model, the correct rate can reach 98.8%, the recall rate is 88.2% and the missing rate is 11.5%. By constructing a sophisticated regional proposal network to solve the problem of long search time, it can effectively solve the problem of excessively long time and make Fast-RCNN real-time identification possible. Results: The quantitative relationship between percentage of pore volume and pore diameter was obtained and the pore diameter distribution was quantitatively analyzed. The analysis results are as follows: the range of pore diameter is mainly between 1 to 50nm; the peak of pore volume is at 13nm; mesopore (Pore diameter between 2-50nm) is the most developed and accounts for 93.7% of the total volume, and the effect is relatively good. Conclusions: The correctness of this method for various types of Neuromedicine and different types of faults can reach more than 95%, the recall rate is more than 78%, and the missing rate is below 15%. The experiment results show that the analysis results are reliable and the description method is feasible. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088), and Post Subsidy Funds for the Sci & Tech Achievements of Northeast Petroleum University (HBZZJ201603). The Corresponding Author is Jianjun Xu. A new method for bad biology data identification based on enhanced immune-fuzzy c-means algorithm Jinlan Gao, Hao Li, Yubo Duan, Limei Yan, Jianjun Xu College of Electricity Information Engineering, Northeast Petroleum University, Daqing, China Background: The EGSA algorithm is used for the Bad Biology Data Identification is very important. Based on the obtained good initial division, the fuzzy clustering is performed by FCM algorithm to obtain the classification of benign data and bad data. Finally, the optimal clustering number is obtained by COS clustering validity judgment index, and the best clustering result is obtained. It is necessary to investigate influence of EMF irradiated by the high-voltage transmission lines on the health of human and animals. Methods: According to the period for Immune-Fuzzy C-Means Algorithm system, We measured the changes of functions of brain, heart and muscle in animals using biology electromagnetic instruments and methods as well epidemiological investigations for 1400 peoples in Heilongjiang, Jilin, Beijing etc., in which we measured also variations of physiological features of peoples live in domain of HVTL and EMFs. Fuzzy c-means algorithm (FCM)is a classical algorithm to solve fuzzy clustering problems. The method is to attribute the clustering analysis to a nonlinear optimization problem with constraints. Relative to other clustering algorithms,FCM algorithm has the advantages of simple design and wide application range. However, the fuzzy C-means algorithm also has the disadvantage of being sensitive to the initial value of the algorithm and easily falling into local optimum. Gravitational search algorithm (GSA) is a new intelligent optimization algorithm with strong global search ability. Results: It can be discover from test consequence of gathering similar sorts of chart、consequence of Clustering affiliating degree of 3 picture and objective function value curve of 4 picture that based on the bad data of EGSA-FCM algorithm to examine measurement data which are classified as same kind, its result indicate that the method can identify them effectively, in the case of the number of gathering similar sorts K=2, obtaining bad data set which involve measurement data from 5 and 15 number, the number of iterations of the algorithm is 5 times, and the convergence number can be obtained. Conclusions: The new method for detecting and identifying bad data in Immune-Fuzzy system based on EGSA-FCM algorithm is proposed, which the enhanced universal gravitation search algorithm is used for early search, which provides a better initial solution for fuzzy clustering and improves the clustering effect and adopt the COS effectiveness indicator to determine the optimal number of Clustering to get the best clustering results. Acknowledgements: This work was supported by the Post Subsidy Funds for Sci & Tech Achievements of Northeast Petroleum University (HBZZJ201603), and the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Limei Yan. 1College of Information Science and Technology, Beijing Normal University, Beijing, China; 2Beijing Key Laboratory of Digital Preservation and Virtual Reality for Cultural Heritage, Beijing, China Objectives: To detect the anatomical boundary of vessel walls in black-blood MR sequences to increase measurement precision of plaque size. Methods: First, a deep convolution neural network is used to determine the approximate location of vessel walls in order to reduce the time of manual definition. Then an improved level set method incorporating the generalized cyclic structure of vessel walls is proposed to simultaneously track the inner and outer anatomical borders. Results: The proposed method outperformed the statistical shape model in tracking both borders. The statistical shape model was easily distracted by plaques attached to the interior wall and produced high distance measurement error. The proposed method produced better results and achieved much smoother vessel wall shapes, particularly when there was a lesion or fuzzy borders. The cyclic structure constraint is optimal for improving the inter-border performance against distraction from plaques. Conclusions: Qualitative and quantitative results on black blood MR images showed the effectiveness of the proposed vessel wall detection method. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants Nos. 61802020, 61472042), and by Beijing Natural Science Foundation (Grant No. 4174094), and by Qingdao Science and Technology Project (2017-4-3-2-xcl). 1School of Computer, Electronics and Information, Guangxi University, Nanning, China; 2Guangxi Key Laboratory of Multimedia Communications and Network Technology, Guangxi University, Nanning, China; 3Liuzhou Worker\'s Hospital, Guangxi University, Liuzhou, China; Wuwudongbobo@126.com Objectives: Segmentation of abdominal organs from CT scans is always time consuming but significant for surgical planning in clinical routine. We design a Web based platform that enable radiologists to access our program running on server via internet, on which a 3D model can be built efficiently without the need of installing the software on their own PC or mobile devices. Methods: Users firstly adjust the contrast and intensity of a sequenced CT image to make a target organ in an optimal view; then select a proper threshold rank for binarization of the whole body scans; finally, by clicking a point inside an organ, 3D labeling method is applied to the pixels connected to this point. A label number is given to these connectivity as one part of a 3D model, and this procedure will repeat until all the organs are labeled into a result file that can be download from web site to the clients. Our web-based program is written by Xojo which is easy to make web application under BASIC programming language and provide virtual server function when in running mode. Results: FAST-EXTRACT, our semi-automatic method for segmenting an abdominal case with 200 CT scans into 8 labels (heart, liver, kidney, spleen, bone, aorta, vena cava, portal vein) costs about 2hours in average, and with accuracy more than 95%. Comparing to our previous fully-automatic segmentation approach LIVERCAD, which extracts 7 labels (without portal vein) within 7minutes, the overall accuracy increases 8% in average. When the same case is performed using fully manual software- IMAGEVIEW, 50hours is needed to keep the same extraction accuracy of 95%. Conclusions: Comparing three types of segmentation methods for organ labeling, our semi-automatic method is able to reach a high accuracy as manual one while deeply decreasing the operation time. We believe that combining the fully –automatic algorithm LIVERCAD into FAST-EXTRACT would further increase extraction efficiency, and it is possible to come to clinical practice among many hospitals with its web manipulability. Acknowledgements: This work is supported by the National Natural Science Foundation of China (Grants Nos. 81460274 and 81760324); and Natural Science Foundation of Guangxi (Grant No. 2017JJA170765y). Molecular mechanism of total flavonoids from Flatstem Milkvetch Seed on perimenopausal syndrome rat Lihua Cao, Mingsan Miao, Xiumin Li, Pengfei Li, Xiaoyan Fang 1Henan University of traditional Chinese medicine, Henan, Zhengzhou, 4500462The Icahn School of Medicine at Mount Sinai, USA, New York, NY 10029 Objectives: To investigate the effect and mechanism of total flavonoids from Flatstem Milkvetch Seed on perimenopausal syndrome rat. Methods: The perimenopausal rat model was established by ovariectomy (Remove all of the left ovary, Excision of 80% of the right ovary). Continuous administration of 35 d. The horizontal movement distance and standing frequency were measured respectively. In forced swimming test,the 2-4min immobility time in the 4min were detected. The intake of sugar was measured at 24h. The Viscera index was calculated. The levels of E2, LH and FSH in serum of perimenopausal rats were measured by elisa. Immunohistochemistry was used to detect the expression of ER, AR, Bcl-2 and Bax in hypothalamus and pituitary. Pathological changes in uterus, thymus and spleen of rats were examined by HE staining. Results: The rat model of perimenopausal period was successfully prepared. In the model group, the level of opening activity, standing frequency and sugar intake were decreased. The time of forced swimming immobility increased, serum E2 level were decreased, FSH and LH levels were increased. The expression of ER and AR in hypothalamus and pituitary were decreased. The index of uterus, spleen and thymus decreased and occurred severe pathological changes. Total flavonoids from Flatstem Milkvetch Seed can improved the symptoms of perimenopausal rats, increased the distance, standing frequency and sugar intake of the model rats, reduced the swimming time of model rat. Total flavonoids from Flatstem Milkvetch Seed also could increased serum E2 level, reduced serum FSH and LH levels, regulated hormone levels, promoted the expression of ER and AR in hypothalamus and pituitary, increased the index of uterus, thymus and spleen, and reduced pathological changes. Conclusion: Total flavonoids of Flatstem Milkvetch Seed can improve the appearance, biochemical and pathological symptoms of perimenopausal rats. Acknowledgements: This work was supported by the National International Cooperation Base (No., 2016-65) and Central Plains Scholars (No. 162101510003). Keywords: Perimenopausal Syndrome, Total flavones of Astragalus seeds, Pharmacodynamics, Mechanism. Research on the application of VR animation to displaying the medical microcosm—Taking the practice of the digestive system diseases as an example Qianru Wang1, Ruibo Yang1, Feng Jiang2 1Sichuan Fine Arts Institute, Chongqing, 401331 China; 2Chongqing General Hospital, Chongqing, 400014 China Objectives: This project takes the diagnosis and treatment of the digestive system diseases as an example, and combines the clinical medicine and the VR technology. Using the three-dimensional animation, the digestive system is simulated and restored, and the upper digestive system and the microcosm of the small intestine are displayed, so that the audience can really understand the knowledge of the medical popular science knowledge. This paper discusses the development of medicine in the future, studies the new application of the reduction technology of the VR animation in the surgeries, scientific researches, training and education and so on, and achieves preliminary results. Methods: The research is to simulate the digestive system for the three-dimensional modeling into a virtual engine. Combined with the computer programming, using the VR animation technology, display the microscopic digestive system, the digestive process and the pathological changes. Users can enter the virtual human body through the VR device for observation and learning, and can simulate the surgery. The virtual reality technology, or the VR technology, is used in the research. It is a virtual space with the visusion, auditory and tactile and so on, which is simulated by computers, so that the experiencers can have the general feelings of being in his own situation. When the user moves the position, the computer can immediately perform the complex operations to send back the accurate 3D world images to create a sense of presence, thereby creating the interaction. Results: This work fully presents the micro-world of the upper digestive system and the small intestine, realizes the integration of the technology and art, and enriches the display types of the medical science popularization. In the process of displaying the works, the audience really knows the science knowledge of medicine, and is impressed. Among them, the \"micro beauty” or the \"scientific beauty” is full of mysterious colors and infinite charm, with the eternal vitality and the expressive force. Looking at our digestive system from the microscopic state, you will find many different places that you know differently, such as a pointed tongue, and the intestines full of the villis and so on. These tiny parts are beautiful that the public does not know. It is hoped that this work will not only give the audience the micro-morphology and the movement laws of the science popularization of the digestive system, but will also bring the audience the enjoyment and experience of beauty. Conclusions: Based on this project, we will continue to do further researches in other areas of medicine, mainly focusing on cells, molecular structures, genes and other micro-world. Moreover, the combination of the VR technology and the popular science industry can produce rich derivatives, VR videos and VR games and so on. The development of the technology is enriching the types of art and making it more convenient to excavate the excellent cultures in every industry. Foundation project This paper is the periodical research result of the scientific and technological research project of Chongqing Education Committee, \"Innovation Research of Medical Clinical Application System Based on the VR Animation Technology: Taking the Creation of Gastroscopy Simulation Operating System as an Example” (No. KJ1708190). 1Guizhou Provincial People\'s Hospital, Guiyang, China; 2Key Laboratory of Advanced Manufacturing Technology, Guizhou University, Guiyang, China Objectives: Congenital cystic adenomatoid malformation is a rare congenital malformation first described by Ch\'In-Tang in 1949, and is a typical clinical manifestation of recurrent infection of the lower respiratory tract. Congenital cystic adenomatoid malformation is usually caused by excessive production of the end of the fine bronchial structure without alveolar differentiation and marked dilatation of diseased lung, it is a benign lesion and most effective treatment is surgical resection of the diseased lung tissue, and usually the prognosis is good. Here we report a rare case of congenital cystic adenomatoid malformation in a Chinese boy. Methods: A 5-month-old boy was admitted to our hospital with intermittent coughs and a 10-day history of fever. His parents reported that he had a medical history of nonproductive cough since his birth, but there was no cyanosis or convulsion, and he had recurrent respiratory infections. He had been treated in many hospitals. Results: At the admission, his body temperature was 37.1°C. The lung auscultation revealed fine rales and wheezing in both lung bases, but other physical examination results were unremarkable. Enhanced computed tomography (CT) of the chest revealed consolidation of inferior lobe of right lung (PanelA, B and C, white arrows) and lots of cystic lesions which contained air (PanelA, B and C, red arrows) or fluid (PanelA, B and C, blue arrows). A diagnosis of congenital cystic adenomatoid malformation was made. After a resection of the inferior lobe of right lung under the Thoracoscope, the boy recovered well and was discharged. Postoperative histopathological examination of the lesion showed alveolar wall covered with columnar cells (PanelD, green arrowheads) and smooth muscle on the alveolar wall (PanelD, black arrowheads), which supported the early diagnosis. At the two-month follow-up, the boy had a good condition. Conclusions: The current case shows the value of surgical resection for the diagnosis and treatment of congenital cystic adenomatoid malformation. Acknowledgements: This work was supported by the grant GZSYQN[2015]09 from the Youth Science Foundation of Guizhou Provincial People\'s Hospital; the grant QKHJC[2016]1532-3. Impact of an E-health intervention program on preoperative anxiety and depression in patients with microbial flora transplantation Qu Shen1, Kexian Liu1, Jianquan He2, Xu Zang3, Yanwei Rao4, Xiaoqing Li1, Yang Liu1, * 1Department of nursing, Medical College, Xiamen University, Xiamen, China; 2Department of Gastroenterology, Zhongshan Hospital Affiliated to Xiamen University, Xiamen, China; 3Rehabilitation Department of Zhongshan Hospital Affiliated to Xiamen University, Xiamen, China4Department of Critical Care Medicine, Jilin Province People\'s Hospital, Changchun, China*Address Correspondence to Dr. Yang Liu: Assistant Professor, 204 Aili Building, Medical College, Xiamen University, Xiangan South Road, Xiangan District, Xiamen, Fujian, China, 361101 Objectives: This article gives an account of a project aimed at ameliorating the preoperative anxiety and depression in patients with microbial flora transplantation by means of an E-health intervention program. The goal was to evaluate how E-Health performs against to traditional treatment. Methods: Based on previous formative research, an E-health interventionprogram including weekly articles about microbial flora transplantation knowledge three times a week via a popular Chinese social media app WeChat was developed and implemented. A total of 100 microbial flora transplantation patients were included in control and intervention groups from June 2017 to February 2018. The patients in intervention group participate in an E-health intervention program, but the control group did not. The intervention lasted for 7days, and all participants’ anxiety and depression were assessed with self-rating anxiety scale (SAS) and self-rating depression scale (SDS) before operation. Results: At base line, the score of SAS and the score of SDS were similar among patients with microbial flora transplantation in the intervention and control group. During the intervention period, compared to the control group, patients in the intervention group had more scores in scores of SAS (t=2.717, P=0.008) and the scores of SDS scores (t=3.054, P=0.003). Moreover, the acceptance of microbial flora were significantly higher in intervention group (χ2=0.577, P 0.01) than which in the control group. Conclusions: This pilot study demonstrates that E-health intervention program could significantly ameliorate the anxiety and depression before fecal bacteria transplantation for patients with microbial flora transplantation. Our data indicates that an E-health intervention is feasible and acceptable to this population. The data were collected through this pilot study will inform the future designs and implementations of E-health interventions in this population. A well-designed mobile rehabilitation app for monitoring and managing patients with microbial flora transplantation can supplement or replace traditional education programs and achieve improved patient health outcomes. Thus, this program is worthy to spread and practice in the clinic for the future. Objectives: As an important nanomaterials, gold nanoclusters (AuNCs) are have been applied in many fields, such as food service and the biology. The presence of AuNCs will be inevitably released into the environment with unknown consequences. Although a few studies have suggested the potential toxicity of AuNCs to terrestrial plants, the lack of standard phytotoxicity evaluation induced by nanomaterials also gave inconsistent results. Therefore, it is an great demand to establish a reliable technical method for phytotoxicity of AuNCs. Methods: We synthesized a kind of gold nanocluster consisting of only tens of atoms to break through the size limitation of plant absorption. Subsequently, Vicia faba seeds were transferred into AuNCs solution with different concentrations to fully interact with each other. The effects of mutagenicity test of Vicia faba root-tip cell micronucleus were demonstrated. Seeds germination percentage, phenotypic morphology of seedlings, and analysis of plant tissue elements, plant physiological status were investigated. Results: When the Vicia faba root tip cells were exposed to AuNCs with conventional concentration in experiments for several hours, micronucleus rates were significantly higher than that in the control group (P 0.05), and the pollution index was even greater than 3.5. By further study we found that the phytotoxic effects of AuNCs were concentration dependent. Consistent with this, the Vicia faba seeds growth was significantly inhibited exposure to AuNCs solution. The contents of ascorbate peroxidase and malonaldehyde increased gradually with the increasing of the time and the concentration of AuNCs solution. Conclusions: In summary, this study provided proof of principle of mutagenicity test of Vicia faba root-tip cell micronucleus on phytotoxicity assessment of AuNCs, which showed the great potential on investigation the mechanisms of the toxicity effects of artificial nanomaterials on plant growth and development. Acknowledgements: This research was financially supported by the National Natural Science Foundation of China (31870486), the Natural Science Foundation of Jilin Province (No. JJKH20170657KJ), (No. JJKH20181186KJ) and the Natural Science Foundation of Changchun Normal University ([2016]06). Research on the restoration application of the VR animation technology in the gastroscopic simulation operating system Qianru Wang1, Zidan Tan1, Binbin He1, Feng Jiang2 1Sichuan Fine Arts Institute, Chongqing, 401331 China; 2Chongqing General Hospital, Chongqing, 400014 China Objectives: This project combines the clinical medicine with the VR animation technology, to develop a gastroscope simulation operating system with the integration of the reality and the virtual reality. The operation process of the gastroscopy and the minimally invasive surgery is restored by the VR technique, and the immersion gastroscopy simulated operation experience is achieved with the assistance of the VR equipment. Methods: Through the combination of the VR animation technology and the clinical medicine, we can create a cross-disciplinary platform, and on this basis, make up for the shortcomings of the traditional medical education. By using the VR equipment, the experiencers can simulate the operation of the immersion gastroscope simulation system, so that the medical workers, the medical students and the general public can better understand the medical knowledge and achieve the purpose of popularizing the medical knowledge, which is conducive to the medical education and improving the medical literacy of all the people. Results: The simulated gastroscope operating system created by the virtual reality technology generates a virtual scene by using the computer and other equipment. It is a kind of the real operating system which integrates the multi-source information and the interactive three-dimensional dynamic scenes and simulation. It allows the users to produce the visual perception, the auditory perception, the tactile perception and the motion perception and so on, thus immersing in the environment of the gastroscopy. Users have an interactive experience with the maneuverability of the objects in the simulated environment and the feedback from the virtual environment. Conclusions: The gastroscopy simulation operating system is a new technology produced in the collision of the VR animation technology and the medicine, and the combination of the two will not be limited to this. With the gastroscope simulation operating system as the starting point, the combination of the VR technology and the clinical medicine will continue to expand in the brain, heart, lung, intestine, nerve and other fields. With the development and improvement of the VR-based medical clinical application simulation system, this research will bring great convenience to the clinical medicine and serve the public better. Foundation project This paper is the periodical research result of the scientific and technological research project of Chongqing Education Committee, \"Innovation Research of the Medical Clinical Application System Based on the VR Animation Technology: Taking the Creation of the Gastroscopy Simulation Operating System as an Example” (No. KJ1708190). Design and application of the hospital payment service platform based on the UnionPay business finance Ao Yang, Jintao Pan Dongguan Third People\'s Hospital, Dongguan, Guangdong, 523320 Objectives: Since its founding, the UnionPay has always regarded the concept of \"integrating into the society and facilitating the masses” as the key to its own business. Therefore, in response to the call for people\'s livelihood, UnionPay has innovated a convenient payment platform to meet the demands of the masses. In the medical field, as an important area of people\'s livelihood, the commercial and financial activities of UnionPay use the patients’ medical cards in the form of the self-service recharging to build a good way for both doctors and patients to pay, so that the entire hospital payment link is greatly optimized. This paper will start with the value analysis of the hospital payment service platform based on the commercial finance of Union-Pay, combined with the design elements of the hospital payment service platform based on the commercial finance of UnionPay, so as to explore the application mechanism of the hospital payment service platform based on the commercial finance technology of UnionPay. Methods: Design and analysis on the hospital payment service platform based on the UnionPay business finance technology. Results: Considering the overall characteristics of the current information age, the medical payment platform based on the commercial and financial technology of UnionPay is gradually improving. This payment method not only replaces the traditional financial payment method, but also realizes the deep integration between the financial management and the technology of the times. At the same time, the reasonable use of the payment service platform will also optimize various resources of the hospitals themselves. Conclusions: In today\'s information age, the Internet finance has become an important part of the public consumption. The hospital payment service mechanism based on the commercial and financial technology of UnionPay will adapt to the demand of the modern hospital transformation on the basis of adapting to the public consumption habits. Therefore, after the construction of a new medical payment platform, through the construction of the multi-channel payment methods, with the improvement of the efficiency of the hospital management as the core, through the integration of the hospital financial system, we can greatly enhance the level of the hospital development. Of course, in building a hospital payment service platform, in addition to selecting the appropriate design elements to meet the necessary system functions, we should also ensure its security, so that the advantages of this payment technology can be better played and presented. Vertical seal evaluation epidemic case-control study in different period of Huhenuo Race, Hailar Basin, Northeast China Mingwang Zhan1, Dianjie Sui1, Yonghe Sun2, Lanzhu Cao3 1College of Mechanical and Electrical Engineering, Guangdong University of Petrochemical Technology, Maoming, China; 2College of Geosciences, Northeast Petroleum University, Daqing, China; 3PetroChina Huabei Oilfield Company, Renqiu, China Background: In this paper, the vertical seal evaluation in different period of epidemic case-control study of Huhenuo race, Hailar Basin, Northeast China, was discussed. Methods: Used the theory, when the Huhenuo race enters the stationary period, Huhenuo race stops grow. If the relationship between Different Period of Huhenuo race, Hailar Basin, Northeast China is shown, the Huhenuo race has not completely offset the caprock from the microscopic angle, the Huhenuo race is vertical sealed. If the relationship between Different Period of Huhenuo race, Hailar Basin, Northeast China is shown like 4c-4d, the Huhenuo race has completely offset the people of Northeast China from the all China. From the above, the active period of Huhenuo race is coincided with the massive expulsion period of hydrocarbon source rocks. The Huhenuo race has undergone both active and stationary period in the expulsion of hydrocarbon. Therefore, Huhenuo race is an important fault in Huhehuoren structure and plays an important role in control the accumulation of Northeast China. Results: The epidemic case-control study in Different Period is vertically sealed at 1˜3, 5, 7, 13 and 15 in active and stationary period. At 8, 12 and 14, the active period is unsealed, but the stationary period is vertically sealed. At 4, 6, and 9˜11, in active and stationary period are both unsealed. Conclusions: The vertical seal property of Different Period 5 in active and stationary period plays an important role in the control of Huhenuo race. The epidemic case-control study is good near the parts where the fault is sealed in both activity period and stationary period. And the epidemic case-control study is good near the parts where the fault is unsealed in activity period but sealed in stationary period. Acknowledgements: Supported by the National Science and Technology Major Project of China (2016ZX05007-003, 2016ZX05006-005), and the National Natural Science Foundation of China (41872157, 4157212, 41802138), the author would like to acknowledge the support of Northeast petroleum university, School of earth sciences, Daqing, China, and also wish to thank Dr. Fu, Professor Zhan, Lecturer Zhao and Dr. Shi for their help and encouragement to do this research. Preparation of environmentally friendly NF base oil for Micro medical robot lubricant fluid Yuxue Sun, Xiuyu Zhu, Jingyuan Zhao, Dianjie Sui, Jianjun Xu College of Petroleum Engineering, Northeast Petroleum University, Daqing, China Background: According to the requirements of environmental protection, at the same time meet the requirements of human absorption, we prepared a new (patented) environmentally friendly base oil, namely, NF oil, use for Micro medical robot lubricant fluid. Methods: Catalytic cracking desulfurization is essentially the carbon-sulfur bond in the organic sulfide molecule being broken by the catalyst (acidic condition) to form hydrogen sulfide, changing the original form of the sulfur, after which it is removed. The stock oil forms gas, oil, and coke during the catalytic cracking, and some coke is deposited onto the surface of the catalyst. As the carbon deposition increases, the activity of the catalyst decreases, eventually leading to a decrease in the desulfurization rate. At this point, regeneration of the catalyst is carried out, meaning that the original reaction is stopped, and the steam mixture is used to burn the carbon deposit, thereby restoring the activity of the catalyst. After fractionation, hydrogenation, desulfurization and aromatization, cooling, and other processes, the aromatic content of the final oil was less than 0.5ppm and the sulfur content was also less than 0.5ppm. This means that the quality of this oil was consistent with the accepted standards. Results: NF oil is obtained from stock oil by fractionation, hydrogenation, desulfurization, dearomatization, cooling, and other processes. NF oil is composed mainly of C12-C22 hydrocarbons, with low impurity contents, small specific gravity, and easy degradability. The LC50 value of treated NF oil is greater than 1000000mg/l WSF, and its toxicity is less than 1/10 of that of other medical base oils. Conclusions: We evaluated NF oil systematically in the laboratory, with the results indicating that the properties of the studied NF oil being clearly superior to those of other medical base oils. It is the preferred base oil for preparing environmentally friendly Micro medical robot lubricant fluid systems because of its high permeability, strong rejection capability and sealing characteristics, minor dynamic filtration, and insignificant pollution. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088). This work is supported by the laboratory of Oil and gas layer protection of the Northeast Petroleum University. The Corresponding Author is Dianjie Sui. Research on fault diagnosis of bearing in vascular motor based on evolutionary deep learning Aihua Xu1, Yang Du1, Guiquan Duan2, Lin Huang1, Jianjun Xu 1College of Electrical Information Engineering, Northeast Petroleum University, Daqing, China; 2The Planning and Design Institute of the NO. 4 Oil Production Plant, Daqing Oilfield, Daqing, China Background: In the research field of surgical robots, the research of micro-drive motors for nondestructive surgery has become one of the hotspots. In this paper, a fault diagnosis technique for bearing of an vascular motor is introduced based on the deep learning of evolutionary neural network. Methods: (1) Optimizer and learning rate selection. Choosing the right optimizer and the optimal learning rate in the CNN network construction process have a great impact on the speed and accuracy of the model training. (2) Selection of the number of convolution layers and the number of convolution kernels. In the process of learning the internal features implied by the original signal samples, CNN network is an important guarantee to improve the classification accuracy rate by constructing a reasonable network level and selecting the appropriate number of neuron. (3) Selection of batch sample size. In the process of deep learning training, in order to achieve the best network training and classification effects, the constructed model requires a large amount of sample data for training. Results: The SDA algorithm, which is another classification algorithm in deep learning, has an accuracy rate of up to 95.8%. BP-NN and SVM are shallow structures, and the diagnostic accuracy of the two cannot be increased with the increase of the number of samples, which limits the improvement of their respective performance. At the same time, the CNN network recognition rate fluctuation range applied in this paper is small, which further proves the possibility of the network\'s one-dimensional time series data classification, and has good robustness and good generalization ability. Conclusions: The algorithm has good robustness and good generalization ability, and can maintain a high fault diagnosis rate for different load types. At the same time, the method still has some shortcomings: Compared with the traditional fault diagnosis algorithm, when the convolutional neural network is applied to the one-dimensional time series data classification, the training time is longer; the parameters of the network model need to be compared through multiple experiments to obtain relatively optimal results. Therefore, it is necessary to further study the ability of the network to be used for adaptive parameter selection in fault diagnosis, and further improve the computational efficiency. Acknowledgements: This work is Supported by Northeast Petroleum University Innovation Foundation for Postgraduate (JYCX_CX06_2018), and the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Lin Huang and Jianjun Xu. 1Department of Stomatology, Longgang Central Hospital of Shenzhen, Shenzhen, China; *Corresponding Author: Yongbo Gao, E-mail: gaoyongbo2007@sina.com Objectives: With the development of digital technique, real-time navigation system was applied in the area of dental implant surgery. The aim of this paper was to evaluate clinical effect of real-time navigation system applied in the dental implants surgery in the esthetic area. Methods: 10 patients with missing teeth in the esthetic area were agreed to receive dental implant surgery by real-time navigation system. The individual fixations were made based on the cast, and biting plates were put on the fixations. The patients with the individual fixations and biting plates were examined by CBCT before surgery. CBCT images were transferred to navigation system to simulate three-dimensional position of dental implants. After high speed hand piece and biting plates matched with the system,dental implant surgery with real-time navigation system were began. With guidance of the system, dental implants were inserted into the bone. Patients were examined by CBCT after surgery. Three or six months later, final crowns were connected to the implant by the abutments. Results: CBCT images after surgery showed that three-dimensional position of dental implants were almost the same as simulation before the surgery. The patients were satisfied with final clinical result of the treatment. Conclusions: The real-time navigation system in dental implants surgery help dentists to put dental implants in the right position as simulated before the surgery, but lack of follow up observation and lots of clinical evidence. Acknowledgements: This work was supported by the National Natural Science Foundation of China (No. 31560265), Shenzhen Knowledge Innovation Project Basic Research Program (No. 20160425170127456), Longgang Medical Hygiene Science Technique Project (No. 20160602122129900). Multisource transfer learning based lymph node detection method via convolutional neural network Yanjun Peng1,2, Yingran Ma1 1College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, China; 2Shandong Province Key Laboratory of Wisdom Mining Information Technology, Shandong University of Science and Technology, Qingdao, China Objectives: In recent years, analysis of medical images exploiting deep learning model, especially convolutional neural network (CNN), has caught attention of many researchers. In practice, a disadvantage of training a CNN from scratch is that a large amount of labeled data is indispensable, which is adverse for biomedical image processing. To address this gap, one of the major techniques that successfully employ CNNs to medical image analysis is transfer learning technique. It is however prone to hand-craft method. One should decide how or which (layer) of the deep transferred features are extracted or frozen/fine-tuning for specific target tasks by many refined experiments. To this end, based on CNN and point-wise (1×1) convolution, we present a novel multisource transfer learning model, which improves the feature transferring procedure in a learnable way for the target task, i.e. lymph node detection. Methods: Through stacking a new devised 1×1 convolution based CNN block, we propose a novel multisource transfer learning approach for lymph node detection. The mechanism behind is straightforward, where the 1×1 convolution is used to adaptively fuse multisource transfer learning knowledge for the target task. Concretely, 1×1 convolutional operation is implemented after pre-trained convolution layers in order to combine the transfer information, replacing the lengthy fine-tuning procedure. To prevent over-fitting and reduce model complexity, we utilize sparse coding algorithm on the pre-trained convolution kernels to extract the corresponding over-complete filter bases layer by layer. Then the 1×1 convolutional blocks are equipped with these transferred \"meta-features” to adaptively generalize them to the target task. At last, a 2-phase training strategy based on sparse connection and dropout technique is devised to improve model feasibility. We use six multisource databases: one natural image dataset, four texture databases and one medical image dataset. And CNN-M model pre-trained on the six databases separately is utilized as feature extractor. Results: The effectiveness of our method is verified on lymph node (LN) dataset: 388 mediastinal LNs labeled by radiologists in 90 patient CT scans, and 595 abdominal LNs in 86 patient CT scans for false positive reduction task. Our method demonstrates sensitivities of about 85%/71% at 3 FP/vol. and 91%/82% at 6 FP/vol. in mediastinum and abdomen respectively, which compare competitively to previous methods. Conclusions: In this paper, we propose a novel multisource transfer learning based method for CT lymph node false positive reduction task. Multisource transfer learning via convolutional neural network presents its considerable feasibility and potential. And the point-wise convolution technique could be used not only to fuse information from different channels as some previous works shown, but to favorably transfer knowledge from multisource databases. Acknowledgements: This work is supported by the Natural Science Foundation of Shandong Province under Grant No. ZR2017FM054, the National Natural Science Foundation of China under Grant No. 61502279, the National key research and development project of the Shandong Province under Grant No. 2016GSF120012, and by Special Project Fund of Taishan Scholars of Shandong Province, Leading Talent Project of Shandong University of Science and Technology. 1School of Automation, Baotou Light Industry Vocational Technical College, Baotou, China; 2Ruixin College, Beijing Institute of Technology, Beijing, China; 3College of Electrical Information Engineering, Northeast Petroleum University, Daqing, China Background: In this paper, a palm type multifunctional health state detector for household use is developed, which is based on a 32-bit high-performance low-power microprocessor MSPK 430. Focusing on the daily application of household users, according to the daily needs of this subject to achieve five tests: blood pressure, pulse, temperature, breathing, walking rate. Methods: According to the period for RUSSIA FUZZY Markov palm type Chain system, the hardware components of the home multi-functional health state detector mainly include sensor circuit, filter circuit, comparator circuit, 128-64 LCD display circuit, keys, USB interface circuit of regulated power supply, clock circuit, data storage circuit and air pump drive circuit. The software part is programmed in C language, and the compiler environment is IAR 3.2. The program mainly includes DS18B20 temperature acquisition, ADC0802 data reading and calculation, 128-64 LCD driver, key reading, reading and calculation of respiratory and pulse signals. Results: It mainly realizes the measurement and calculation of five basic signs of human body, and displays the information on palm type LCD. The collected data is stored in data memory. When the tester and computer are connected through USB interface, the stored data can be read from palm type PC. In addition, through the continuous revision of hardware and software design, the system error of this design is reduced to a great extent, so as to improve the accuracy and sensitivity of detection, and make the whole system achieve the purpose of simplicity, accuracy and anti-interference. Conclusions: The apnea of the monitored object of the palm type household multi-functional health state detector during sleep has the function of intelligent detection and alarm, which makes the apnea patients get timely treatment and greatly reduces the probability of sudden death in sleep. Acknowledgements: This work is supported by the project of Nature Scientific Foundation of Heilongjiang Province (Grants No. E2017011). The Corresponding Author is Jianjun Xu. 1Dept. of Computer Science, Graduate School, Sangmyung University, Seoul, South Korea2Dept. of Computer Science, Sangmyung University, Seoul, South Korea Background Recognizing objects from photographs and classifying them into predefined classes is a challenging topic in machine learning. Recently, techniques based on convolutional neural network (CNN) accelerate the progress of the recognition techniques [1]. Therefore, the domain of recognition is extended to various practical areas such as food and dietetic items. We propose a CNN-based food recognition framework that improves the performance of recognition and provides dietetics information for users through photo-based interface. We collect 6K photographs of Korean foods of 18 categories and use them for training and testing our classifier. Methods: We design a recognition model for food by employing a deep convolutional neural network. We employ VGGNet structure as the first part of our recognition model [2]. The pre-trained parameters of the original VGGNet are further trained through the food photographs from our database that is composed of more than 6K food images. To visualize the proof of evidence and to improve the accuracy of recognition, we design the second part of our model as a fully convolutional layer, which is composed of 512 filters with relu functions. The result of the first part is further processed into three layers of the fully convolutional layers and fully connected layers in the second part. The collected 6K food photographs are grouped into training (70%), validation (15%) and test (15%) subgroups. Results: We implement our model on a PC with Pentium i7 CPU, 32 GByte main memory and nVidia TitanX GPU. Our model is developed based on a pretrained VGGNet-19, which is implemented using Tensorflow and CUDA libraries. Our model classifies food images into 18 categories and the dietetic information for the food in each category is presented for users to advise their own purposes. For a convenient interface, we implement an image capturing module on a mobile phone and the image recognition module on a server with Linux. These modules communicate via node.js environment. Our model shows 96.50% accuracy for training dataset, 94.11% for validation, and 93.09% for test. Conclusions: We have developed a food recognition model using a deep CNN and proved the excellent accuracy for recognizing foods of 18 categories. We also plan to extend our model by devising a hierarchical structure that recognizes food in two stages, which will improve the accuracy and to apply other CNN models such as DenseNet and ResNet for our recognition model. Acknowledgement: This paper is supported by Sangmyung Univ. Research Fund at 2016. References: [1] K. Simonyan and A. Zisserman, \"Very deep convolutional networks for large-scale image recognition. arXiv print arXiv:1409.1556, 2014. [2] A. Krizhevsky, I. Sutskever, and G. Hinton, \"Imagenet classification with deep convolutional neural networks”, In Proc. of Advances in Neural Information Processing Systems, 2012, pp.1097-1105, 2012. The influence study on the development of medical mode for the form of hospital buildings An Li, Ruixin Wang Xi\'an University of Architecture and Technology, College of Architecture, Xi\'an Shaanxi 710055, China Objectives: To explore the influence study on the development of medical model for the hospital architectural form Methods: using the existing types of hospital buildings in various provinces of China and the future development plan, this paper analyzes the function, centralized layout, interior and future development direction of hospital buildings from the mechanical and biomedical point of view. Results: With the development of mechanical and biomedical models, the functions of hospital buildings are gradually clarified, the layout of buildings is centralized, the internal equipment are more intelligent (elevators, transportation and air supply), and the hospital is developing towards bio-psycho-social medical model, material and digital medical environment. Conclusions: Medical Model Promotes the Development of Hospital Architectural Form towards Centralization, Intelligence and Intelligence from Mechanical and Biological Perspectives Acknowledgements: Supported by a project grant from +++++ (Grand No. 111111). Evaluation of residual fat after injection of oil in high water cut stage of human-bone Xiaozhe Yin1, Junhui Guo2, Jianjun Xu3 1School of Computer and Information Technology, Northeast Petroleum University, Daqing, China; 2E&D Research Institute, Daqing Oilfield Company Limited, Daqing, China; 3School of Electrical Information Engineering, Northeast Petroleum University, Daqing, China Background: Since ATP is often attached to protein cells, we have established a new mechanism and theory for energy transfer of bio-fat released by hydrolysis of ATP molecule, that is, ATP molecule transfers along the direction of protein cells and leads to structural changes of protein cells. It is necessary to study their furnace changes, which means that they are widely used in physical medicine. Methods: Forty-four patients with protein cell disorders admitted from January 2018 to March 2018 were randomly divided into two groups (21 cases in mild preeclampsia group and 23 cases in severe preeclampsia group). 27 male patients in the same period were selected as control group. HQE staining, immunohistochemistry and Western blot were used to detect the expression of HOXB4-7. Compared with the control group, the expression of HOXB4-6 protein was compared with that of the control group, and the regulation of ATP expression was analyzed. These changes are studied using the new Pang\'s bioenergy transport theory and protein cell structure. Firstly, effective displacement radius is studied on the basis of exploration wells, and the influence of reservoir physical properties on displacement radius is identified by reservoir heterogeneity index. Results: The low expression of HOXB4 in ATP patients provides a clinical basis for studying the etiology, pathogenesis, prevention and treatment of ATP-induced protein cells. Field application shows that, compared with engineering analysis, the work efficiency is improved hundreds of times, and the accuracy is about 95%. Therefore, Pang\'s theory system is very correct. Conclusions: Evaluation of residual fat after injection of oil in high water cut stage of human-bone, we use residual fat to explain and elucidate clearly protein cells the mechanisms and characterizations of changes of protein cells many biological and physiological phenomena in physical medicine. Acknowledgements: This work was supported by the Post Subsidy Funds for Sci & Tech Achievements of Northeast Petroleum University (HBZZJ201603). The Corresponding Author is Jianjun Xu. Prevalence and risk factors of overweight and obesity among school-age children aged 8 to 12years in urban China Junping Zhang1, Yunyun Hu2, Shilin Deng3 1College of Physical Education, Hubei University of Technology, Wuhan, China; 2College of Physical Education, Hubei University of Technology, Wuhan, China; 3College of Physical Education, Wuhan University of Technology, Wuhan, China Abstract: To understand the prevalence of overweight and obesity among 8-12year old school children in China and analyze the causes and risk factors. Methods: From December 2015 to June 2016, a total of 5000 questionnaires, 4466 questionnaires and 4386 valid questionnaires were distributed in 6 cities of Shanghai, Shijiazhuang, Lanzhou, Zhengzhou, Wuhan and Shenzhen. That is to say, a questionnaire survey was conducted among 4386 primary school students in 6 cities. Results: The detection rate of overweight and obesity was 19.76%, the overweight rate was 11.03%, the obesity rate was 9.24%. The distribution of overweight and obesity was different in gender, age and region. The overweight and obesity rate of boys (22.78%) was higher than that of girls (16.54%) (P 0.01); the overweight and obesity rate was over 10years (18.54%) higher than those below 10years (17.36%); overweight and obesity rates were higher. It is directly proportional to the regional economy. Multivariate analysis showed that the risk factors for overweight and obesity in children were: physical activity was insufficient (OR=4.178, 95% CI=3.487-5.001), boys (OR=1.227, 95% CI=1.007-1.435), no meat eating (OR=0.723, 95% CI=0.548-0.949), no favorite snacks (OR=1.346, 95% CI=1.049-1.741), and slow eating speed (OR=0.354, 95% CI=0.221-0.587) and diet structure (OR=0.557, 95% CI=0.325-0.966) helps to reduce the incidence of overweight and obesity among school-age children. Conclusions: Intervention strategies aimed at promoting the daily life habits, physical activity level and healthy diet of school-age children are the key and focus to prevent and intervene overweight and obesity. Acknowledgements: This work was supported by the National Natural Science Foundation of Hubei, China (Grants No. 2014CFB861). Effect of Tai Chi on the rehabilitation of neck type cervical spondylosis in the elderly Youlin Qian, Shijie Liu, Lin Wang Dept. of Physical Education, Wuhan University of Technology, Wuhan, China Objectives: Neck Type Cervical Spondylosis has become a common chronic disease which threats to individuals’ health. It causes of cervical osteoarthritis, cervical nerve root syndromes and synthetic hyperplasia. Tai Chi originated in ancient China, which characterized by gently physical activities, mind-body exercises, in order to achieve the purpose of sports, fitness and mediation. Its mild-to-moderate exercise intensity is suitable for individuals with medical conditions. There are several researches show that tai chi is good for the flexibility, stability of the cervical spine. And it also could promote the blood circulation of the neck and relieve the pain. Therefore, this study wants to provide a better preventive and health care mechanism for patients with Neck Type Cervical Spondylosis from the perspective of \"class traction” caused by practicing Tai Chi. Methods: 50 participantswho is by Neck Type Cervical Spondylosis were randomly divided into Tai Chigroup and control group. There were 13 males and 12 females in the Tai chigroup, themean agewas62.55±8.82years.The average disease duration was 6.71±3.68months. 10 males and 15 females in the control group, themean agewas63.21±8.39years, mean duration was 7.12±3.78 months. The control group with conventional treatments, the Tai Chigroup with Tai Chi and conventional treatments,with2 times a day, 30minutesper time, comparative efficacy after 12weeks. Results: The pain (VAS) score of the patients in the Tai chi group was significantly lower than that of the control group (P 0.01). After the intervention, the difference was statistically significant in the Cervical Spine X-ray(P 0.05). The cervical vertebra curvature of the cervical spine was improved, and the curvature changed from anti-tension to straight. Conclusion: Tai Chi exercise as an adjunctive treatment may potentially improve smooth ventilating blood, smooth emotion, strong bones and strong bones of Cervical Spondylosis patients as well as quality of life. Tai Chi exercise could be tentatively prescribed for Neck Type Cervical Spondylosis in combination with the conventional rehabilitation program to quicken the process of recovery. Acknowledgements: This work was supported by The Ministry of education of Humanities and Social Science project (17YJA890025); The Fundamental Research Funds for the Central Universities (WUT: 2018VI014). 1Information Technology Department, Higher college of Technology, Muscat, Sultanate of Oman; 2Department of Instructional and Learning Technologies (ILT), Sultan Qaboos University, Muscat, Oman; 3SCMS Centre for Robotics, SCMS School of Engineering and Technology, Ernakulam, Kerala, India Background: During the learning process, whether students remain attentive throughout the session influences their learning capability. If teachers can identify whether students are attentive they can be notified to remain focused, thus resulting in improving their learning capability. Traditional methods require, teachers observe students’ facial expressions to identify whether they are attentive during the session. However, this method is often inaccurate and increases the burden on teachers. The number of Alzheimer\'s patient and resulting deaths is increasing every year mainly due to delay in the early detection so as to take necessary measures and treatments to overcome it. With the development of electroencephalography (EEG) detection tools, a window has opened for developing an effective equipment to aid this cause. Objective: The aim is to develop a device that can check the concentration level, memory capacity and arithmetic levels of students and for the early detection of the Alzheimer\'s disease. Teachers can put on remedial methods for weaker students by offering them with special attention and care. Early detection of Alzheimer\'s disease will help in taking necessary measured to get cured or control the extent of it. Methods: Every subject undergoes a session of trials for computing the subject-specific threshold to assess his/her attention and concentration level to perform specified tasks. The subject will be sitting in an upright posture wearing the EEG device, presented with the some objects or a sheet of patterns. Session has 3 phases such as preparation, concentration and relaxed phases. During idle phase active concentration to object or point on the pattern is avoided. In the concentration phase, user actively concentrated at a specified object or part of the pattern. In Relaxed phase subjects are free to divert their focus from the objects and patterns. Analysis of the data was done to estimate the values of the EEG signal extracted during the active concentration period. The average data of the trials is taken as the threshold. The threshold and the extracted raw data are compared and analyzed. The resulting data is used for the brain activity analysis which would help in the effective detection of Alzheimer\'s in its early stages. Results: Effective EEG signal extraction is achieved. Efficiently analyzed concentration levels of students thereby helping the staffs to analyse their concentration levels and employing necessary measures to enhance the brain activity and concentration skills of the students. In healthcare early detection of the Alzheimer\'s was very supportive to help the patient and provide necessary treatment and care at the early stages itself. Conclusion: A more user friendly, compact and better portable system will be designed such that it could be easily used in education institutions. The easy portable design will also be very helpful in hospitals aiding the patient to move around and interact by wearing the device. Study and design of physical health management system for civil servants in Hubei province of China Qi Luo College of Sports Engineering and Information Technology, Wuhan Sports University, Wuhan, 430079, China Objectives: The purpose of this paper is to construct a complete civil servant health assessment index system (adding psychological and social adaptation indicators), and to take civil servants in Hubei province as the research object, aiming at comprehensive, objective and accurate evaluation of the physical health status of civil servants in Hubei province. On the basis of seizing the factors affecting the health of the civil servants, we searched the opinions of the collective health experts, designed targeted exercise prescriptions, and finally developed and designed the Hubei province civil servant health management integrated system (including database, expert database and website), which support for the practice of promoting the health of civil servants. This project is the need to implement the scientific concept of development, strengthen the quality of civil servants, and build a civil servant team that is lean, efficient, diligent, honest and pragmatic, so that civil servants can understand their physical health and seize the factors affecting the health of the civil servants; They are concerned about their body health condition and improve their health awareness. Organizations at all levels can also understand the physical health of civil servants, thus laying the foundation for the health management of civil servants. In this study, simple and easy-to-act exercise prescriptions are beneficial to civil servants to accept and implement, so that they can improve their physical health quickly and effectively. Civil servants are also representatives of the government\'s image. The application and implementation of this system has important demonstration significance and leading role in promoting the popularization of the National Fitness Program Outline in the whole society. Methods: The main research methods of this paper are as follows: 1. Literature review method Search and inquire about the research literature and policies and regulations of this project at home and abroad, sort out the civil servants’ body health status, existing problems and development direction of civil servant health research, and clarify the objectives and research paradigms of this project. 2. Physique test method Go to Wuhan, Yichang, Shiyan, Xianning, Huangshi, Enshi and other cities to conduct computer software and instrument testing on the physical health of 800 civil servants in 3-5 government agencies, collect data, and evaluate. 3. Expert interview method 10-15 experts (including senior management, physical health experts, and mental health experts of the Civil Service Bureau) are visited. Conduct feasibility evaluation of project implementation and expert evaluation of implementation effects are given. 4. Questionnaire and interview method Questionnaires were used to survey the physical health needs, interests, attitudes and attitudes of 800 civil servants, as well as health and fitness awareness. In-depth interviews were conducted with 15-20 representatives of ordinary civil servants. 5. Mathematical statistics Descriptive mathematical statistics, analysis of variance, and nonparametric tests were performed on the survey results using the SPSS.12.0 software package. 6. System Design Method In order to link the physique health website and the health database with the prescription library background seamlessly, C/S and B/S are combined to develop advantages. C/S mode service model: its operating system adopts Windows Server2000; the database adopts SQL server2000; the development language adopts VC, B/S mode service model. Client PC terminal: the operating system adopts Windows Profession 2000, interacts with the Web Server in embedded mode, and can directly interact with the database, and finally feeds back the generated HTML document to the Web Server. The development language uses the scripting language.net. The technical route studied in this paper is as follows: 1. Preparation Literature review are concluded on physical fitness, health, civil servant competency and exercise prescription, a small sample survey and interviews were given with civil servants in Hubei province. At the same time, experts in the field of physical health research were interviewed. Its purpose is to establish standards for physical fitness testing, and to develop research and development teams that include disciplines in physical education, psychology, and computer science according to the needs of the research. 2. Build an evaluation system Based on the results of the preliminary work, the content analysis method is used to construct the three-dimensional physical health dimension, and its feasibility and effectiveness are tested. Evaluation indicators, tools, methods, procedures, and standards for various dimensions of physical health are written and Test Manual for Three-Dimensional Evaluation System for Physical Health is compiled. 3. Conducting a physical survey Using stratified random sampling method, a representative group and region were selected as the survey objects in Hubei civil servant group for physical health survey. There are 800 people in the preliminary design survey sample. The collected data is analyzed, the status quo and existing problems of civil servant health is understood and the potential causes are understood through structured questionnaires and interviews. 4. Design exercise prescriptions In combination with the status quo of civil servant health and its work and lifestyle, we designed targeted and practical exercise prescriptions, and tested the effects and work of healthy exercise prescriptions on the health promotion of civil servants through on-site experiments. 5. System development Develop a comprehensive management system for the physical health of civil servants in Hubei province, including modules for physical fitness assessment and exercise prescription design. The basic modules are Figure1. Figure1: Basic modules of management system for the physical health of civil servants in Hubei province Results: The main research results of this paper are as follows: 1. A three-dimensional (physiological, psychological, social) evaluation index system for civil servant health was established, and tools and standards for civil servant health testing were established. 2. From the sample of civil servants in Hubei province, the three-dimensional evaluation of the physique was constructed to test and the norm was formed. The status quo of civil servants in Hubei province is analyzed and then forms a research report. 3. Interesting, simple and practical fitness tools and instruments suitable for working methods and lifestyles of civil servants have been developed. A complete civil service fitness exercise prescription is formed, edited and published. 4. The Hubei province civil servant health database and prescription library were built. 5. A comprehensive system for the physical health management of civil servants in Hubei province is developed. 6. The civil servant health network of Hubei province was built. Conclusions: 1. The psychological and social adaptation dimensions are explanatorily introduced in the evaluation of civil servant health, so that the evaluation of civil servant health in Hubei province is more comprehensive and systematic. 2. The designed and developed civil servant health management system can realize the online testing, evaluation, monitoring and management integration of civil servant health. 3. A civil service physique health exercise prescription system established. According to the results of the civil servants physical fitness assessment, the fitness exercise prescriptions suitable for the civil servants are designed to improve the fitness awareness of civil servants and help them form positive and reasonable exercise and fitness habits. 4. Internet technology is used to realize visual, interactive and participatory management of civil servant health management, and promote the scientific and dynamic management of civil servant health. Acknowledgements: This work was supported by 2014 Science and Technology Foundation of Hubei Provincial Department of Education (Grants No. Q20144102, Research on Health Promotion and Management System of Civil Servants in Hubei Province of China). The rehabilitative effect of health Qigong in chronic obstructive pulmonary disease: a comparative study of Liuzijue and Wuqinxi Lin Wang1, Xiaohu Jin1, Zaimin Li2 1Dept. Of Physical Education, Wuhan University of Technology, Wuhan, China; 2College of physical education, Zunyi Normal University, Zunyi, China Objectives: Chronic obstructive pulmonary disease has become a common chronic disease. The dyspnea and declining exercise capacity would result in sedentary, which lead to a dangerous vicious circle of physical functions declining. Backing to normal life is an ideal condition which COPD patients want to. Improving of lung function, exercise capacity, dyspnea are extremely important for COPD patients. Health qigong as an exercise from China, it include physical, mental and breathing exercise. There are a lot of researches show that health qigong has an effectiveness of human beings. As a breath-dominated exercise, Liuzijue is different from Wuqinxi which concludes not only breathing exercise, but also concludes strength, balance and so on. So, the aim of our study is to make a contrast between Liuzijue and Wuqinxi, pointing out which one is better for COPD patients, offering a better exercise for COPD patients to improve their quality of life. Methods: 45 participants were selected from the respiratory department of Hubei Provincial People\'s Hospital and Wuhan Zhongnan Hospital who have been diagnosed with COPD. We utilized a randomized control trail design, with participants assigned in a 1:1 ratio to three groups, control group with usual care (n=15), Liuzijue group with Liuzijue and usual care (n=15), Wuqinxi group with Wuqinxi and usual care (n=15). The objects received a 12weeks, 5 times for par week and at least one hour each time exercise training. Lung function, 6-Minute Walk, fatigue and life of quality would be tested before and after the intervention. We used SPSS23.0 to analysis the test data, finding out the differences. Results: As an effectiveness method for physical health, both Wuqinxi and Liuzijue could be beneficial to lung function, dyspnea and life of quality. And the results showed that Liuzijue was better than Wuqinxi in improving lung function and dyspnea, but the difference is not significant. Conclusions: We can conclude that health qigong could as a rehabilitation method for COPD patients. And the result of health qigong could provide preliminary regarding the value of health qigong for improving COPD patients’ lung function, dyspnea and quality of life. At the same time, it could offer evidence for mind-body exercise in chronic disease. We considered that the reason why there was no significant difference between Liuzijue and Wuqinxi, it maybe because of the insufficient time and amounts. So, we suggest that the intervention duration should be changed into 24weeks or 48weeks in the follow-up study, to observe the different effect of intervention on pulmonary function and quality of life in COPD patients. Acknowledgements: This work was supported by The Ministry of education of Humanities and Social Science project (17YJA890025); The Fundamental Research Funds for the Central Universities (WUT: 2018VI014). Study on the construction and evaluation standard of physical fitness index system for adolescent women tennis players in China Yong Ke1, Siyuan Wang2 1Sanhe International Tennis College, Wuhan Sports University, Wuhan, 430079, China; 2Graduate School, Wuhan Sports University, Wuhan, 430079, China Objective: This paper analyzes the characteristics of tennis sports from a systematic point view, and completes the construction and evaluation standard of the index system of adolescent physical fitness by combining the development needs of adolescent’ physical qualities. Method: The paper uses the methods of literature, expert interview, questionnaire, experimental test, mathematical statistics and logical analysis. Through the expert interview and questionnaire survey, the physical quality index is optimized and screened to test the index data for athletes, SPSS and excel software are used for statistical data analysis, analytic hierarchy process (AHP) and standard percentile method are used to complete the index weight and evaluation standards, the physical quality evaluation index model for China\'s adolescent women tennis players is established preliminary. Result: The conclusions are drawn as follow: (1) This study takes the physical quality of Chinese adolescent women tennis players as the entrance. According to the principle of scientific, directional, relative independence, feasibility and consistency, after the primary selection of indicators and the screening of experts, 5 first-level indexes are obtained, namely strength quality, speed quality, endurance quality, sensitive quality and flexibility quality; There are 14 secondary indicators: grip strength, stationary tennis throw, standing long jump, 1min push-ups, 1min sit-ups, 30-meter run, single stroke sideline skid, 45second forehand swing, 1000m run, 400m run, left line forward sprint run, right line reverse sprint run, sitting body flexion, left and right split. (2) Using analytic hierarchy process (AHP) to establish weights for 5 first-level indicators and 14 level two indicators. The weights of the first-level indexes are strength quality (0.38), speed quality (0.27), endurance quality (0.11), sensitive quality (0.17) and flexible quality (0.07). Level two index weights are grip (0.06), stationary tennis throw (0.42), standing long jump (0.26), 1min push-ups (0.16), sit-ups (0.10), 30 meters run (0.30), single stroke sideline skid (0.54), 1minutes forehand swing (0.16), 1000 meters run (0.66), 400 meters run (0.34), left line forward sprint run (0.5), right line reverse sprint run (0.5), sitting body flexion (0.25), left and right split (0.75). (3) The evaluation standard of individual physical fitness index, primary index level and comprehensive score grade for Chinese adolescent women tennis players were established. The evaluation standard for the first-level indicators are strength quality: Excellent 30.46 points above, good 26.56-30.46 points, medium 14.86-26.56 points, lower 9.96-14.46 points and below 9.96 points for poor; Speed quality: Excellent 22.80 points above, good 19.65-22.80 points, medium 10.20-19.65 points, lower 6.00-10.20 points, poor and inferior 6.00 points below; Endurance quality: Excellent 8.73 points above, good 7.62-8.73 points, medium 4.29-7.62 points, lower 2.81-4.29 points, poor 2.81 points below; Sensitive quality: Excellent 13.60 points above, good 11.80-13.60 points, medium 6.40-11.8 points, lower 4.00-6.40 points, poor 4.00 points below; Flexibility quality: Excellent 6.24 points above, good 5.48-6.24 points, medium 2.44-5.48 points, lower 0.92-2.44 points, poor 0.92 points below. The evaluation standard of comprehensive grades is: Excellent 70.51 points above, good 63.91-70.51 points above, medium 44.11-63.91 points above, lower 35.31-44.11 points above and inferior 35.31 points below. Conclusions: According to the adolescent tennis player\'s body single evaluation standard, we can evaluate the scores and comprehensive grades for the athletes’ physical fitness, and draw the shortcomings of the young tennis players in a certain physical quality, for the tennis coaches to diagnose, reference or evaluate the adolescent tennis players’ physical fitness level and even the competitive level, promote the development of adolescent women’ tennis. Acknowledgements: Wuhan Sports University East Lake Scholars Program Funding. Objectives: A basketball player is a player who can either score or drive a teammate in a game, and who can turn things around at key moments in the game, such as Kobe Bryant of the Lakers, Lebron James of the heat, rose of the bulls, Anthony of the Knicks, etc. For excellent athletes, especially the core players, they should have both good physical quality and good psychological quality, maintain a certain psychological state, and better display the tactical level. Many studies show that when players feel good about themselves, they shoot 79 percent of their shots and 92 percent of their free throws. Shooting 41% from the field and 46% from the free throw line in frustration and panic. However, in the case of low confidence or extremely unfavorable game atmosphere, the shooting percentage of shots and free throws is only 20%, which indicates that the psychological state of players has a great influence on the result of the game. The study choose the male players of the CUBA team participating in the national CUBA division as Subjects. It will compare psychological qualities of core players and non-core players in sports motivation, confidence, coping level, will quality, personality traits, anxiety and leadership, then analyze the advantage of the key players on the psychological quality and its leading role in the game. This study will provide a basis for the cultivation and experience of psychological quality in basketball training and a reference for the selection and training of basketball players, especially for CUBA players. Methods: Subjects: The players of the 38 basketball teams participating in the national CUBA tournament. Among the core players, 80 (69 men and 31 women), average age of 20.32, (SD= 1.02); The non-core players were 100 (58 men and 42 women), the average age of 20.86 was (SD= 1.13), the effective questionnaire was 96, and the effective level was 96 percent. The core players of the Cuba team are determined to determine the overall evaluation of the player\'s ability. The data index of the core players has the appearance rate, the field is all the time, the field of the field, the field is all score, and the evaluation and recognition of the coach of the team. Methods: The research methods adopted in this study include expert interview and questionnaire. Expert interview: The coaches of the 16 teams were interviewed to obtain relevant information about the key players of the CUBA teams and their psychological quality characteristics, providing references for the selection of psychological quality measurement indicators in this study and the cultivation and training of psychological quality of the team members. Questionnaire: In this study, seven major psychological qualities related to basketball training and game performance were identified, including sports motivation, sports confidence, coping ability, quality of will, personality characteristics, sports anxiety and leadership. Instruments: The measuring instruments used in this study include sport motivation scale (Zhang L W, 2001), sports (Vealay, 1986), self-esteem scale athletes cope with scale (Si G Y, 2006), will quality scale (Yin X C, 2001), the WT athletes personality inventory (Qi C Z, 2000), a motion cognitive trait anxiety inventory (Ye P, 2000), the leadership skills questionnaire (Carter, 1983). Results: (1) Sport motivation The average score of CUBA core players’ participation tendency is 9.82, the average score of avoidance tendency is 4.50 and the average score of sports motivation is 5.32. The average of CUBA\'s non-core players’ participation tendency is 9.03, the average of avoidance tendency is 6.54 and the average of sports motivation is 2.49. The difference test showed that the core players CUBA have higher motivation to participate than non-core players, and the avoidance tendency is lower than non-core players. This indicates that core players have higher motivation level than non-core players, and that core players are more willing to participate in and participate in basketball training and competition. (2) Sports self-confidence The average of CUBA core players’ total sports confidence score is 78.16 and 83.02 respectively. The mean of the total score of the athletes with special characteristics of non-core players was 72.75, and the mean of the total score of the state sports was 72.31. The difference test showed that the core players CUBA have high athletic confidence, and their special sports confidence and state level of athletic confidence are higher than non-core players. It can be considered that CUBA\'s core players have higher sports confidence than non-core players. (3) Athletes’ coping ability The average of CUBA\'s core players’ response points to solve problems centrally is 22.34, while the average score for non-core players is 18.92. The average score of core players’ coping with emotions centrally was 23.45, and that of non-core players was 21.38. The average score of core players in evasive coping was 13.45, and that of non-core players was 14.22. The average of core players’ total points in transcendence coping was 10.34, while non-core players scored 9.21. The difference test showed that the core players CUBA have better coping ability in sports. Their coping level of solving problems and coping with emotions are higher than that of non-core players, which indicates that core players have higher ability to solve problems and deal with emotions. (4) Athlete\'s quality of will The average of CUBA\'s core players’ awareness score is 23.56 and that of non-core players is 20.27. The average score of core players’ decisiveness is 25.76 and that of non-core players is 23.46. The average of core players’ toughness score was 28.43 and that of non-core players was 25.64. The average of self-restraint score of core players was 16.44, while that of non-core players was 14.32. The average of the core player\'s total initiative score was 11.34, and the non-core player\'s score was 9.34. The difference test showed that the core players CUBA will have better quality, its self-consciousness, decisiveness, self-control, tenacity, initiative are higher than the non-core athletes, explain the core will quality all dimensions were higher than the non-core athletes, athletes show non-core athletes is higher than that in the core player self-consciousness, decisiveness and self-control, tenacity, initiative will quality, etc. (5) Personality characteristics of athletes The average of the toughness score of CUBA\'s core players is 45.56 and that of non-core players is 43.45. The average of core players’ excitability score was 48.93, while that of non-core players was 46.78. The average of the core players’ total social scores was 49.23, and the non-core players’ scores were 45.54. The average of the core players’ rational total score is 47.45, while the non-core players’ score is 46.45. The average of core players’ self-control score is 46.33, while that of non-core players is 44.34. The average of emotional total score of core players was 52.53, and that of non-core players was 50.45. The average of the core player\'s aggressive total score was 43.34, and the non-core player\'s score was 40.45. The average of core players’ flexibility total points was 49.45, while non-core players scored 47.86. The difference test showed that the core players CUBA are more resilient, sociable, autonomous and aggressive than non-core players, indicating that they are more resilient, sociable, autonomous and aggressive than non-core players. (6) Athletes’ anxiety level The average score of social evaluation anxiety of CUBA core players is 14.78, while that of non-core players is 13.23. The average score of core players’ preparation anxiety is 12.73 and that of non-core players is 11.87. The average score of core players’ performance anxiety was 14.57, and the score of non-core players was 13.56. The average score of core players’ failure anxiety was 6.32, and that of non-core players was 7.34. The average score of core players’ anxiety about each other\'s strength was 8.33, and that of non-core players was 8.94. The average of the injury anxiety total score was 6.45, with non-core players scoring 6.57. The difference test showed that the core players CUBA have higher social evaluation anxiety, competition preparation anxiety and competition performance anxiety than non-core athletes, lower failure anxiety, opponent strength anxiety and injury anxiety. It shows that core athletes are more concerned about social evaluation, competition preparation and competitive performance than non-core athletes, and have higher anxiety level in these aspects. (7) Athlete leadership The core players of CUBA score 4.34 on teamwork, while non-core players score 3.35. Core players score 3.51 on self-understanding and non-core players score 2.45. Core players scored 3.46 in communication, while non-core players scored 2.98. Core players scored 4.08 in decision-making and non-core players scored 2.85. Core players scored 3.95 in leadership, and non-core players scored 2.08. Core players scored above the median 3 points in each dimension. The difference test showed that the core players CUBA have better leadership and are better at teamwork, communication, decision-making and leadership than non-core players. It shows that core players have higher leadership skills such as teamwork, communication, decision-making and leadership than non-core players. Conclusion: The core players CUBA team scored higher on all aspects of mental qualities, superior to other players in sports motivation, confidence, coping, quality of will, character, anxiety and leadership, which makes the core players have the ability to lead the team to achieve goals and common vision, for they play a key role in the basketball training and competition, and lead the team to victory. Suggestion: Players psychological quality has great influence to the success of the game, so coaches in pay attention to the athlete body quality and movement technical level at the same time, also should focus on the players’ psychological quality, scientific and reasonable measurement and evaluation method, on the players’ psychological quality monitoring and psychological training, to effectively improve the whole team players psychological quality, improve the comprehensive quality, improve the level of the team. For non-core players, attention should be paid to the development of their motivation, confidence, quality of will and coping level. In the long - term basketball training, to shape its character; Improve sports anxiety, improve teamwork and communication skills, and lay a foundation for better improvement of basketball skills and the full play of the game. Acknowledgements: This work was supported by the special fund for basic scientific research operation expenses of central universities (Grants No. 2016VI066). Study on mental health monitoring and sports intervention of left-behind children in cross-border ethnic groups between China and Vietnam Gong Chen* Wuhan University of Technology, Wuhan, China, 430070Guangxi Economic and Trade Vocational Institute, Nanning, China, 530021Gong Chen (1980–), Male, born in Hengyang, Hunan province, Associate professor in Guangxi Economic and Trade Vocational Institute, Ph.D. candidate in Wuhan University of Technology, main research: National Traditional Sports. Background: The length of the land border between China and Vietnam has reached more than 1300 kilometers, and there are 12 nationalities identified by China. In addition, the ethnic composition of the Sino-Vietnamese border is relatively complex, and the number of ethnic groups is numerous. Different nationalities have different living habits. People settling down there are known as cross-border ethnic groups. The characteristics of children left behind in the border of China and Vietnam are obvious- poor living conditions, diverse national cultures, lack of attention and deficient education. In recent years, more and more people of cross-border ethnic groups have chosen to go out for work because of the influence of the external environment, and a large number of minors have reluctantly become left-behind children. In order to explore the mental health problem for these children and provide the effective intervention for their mental health development, the author focuses on the children left behind across the border between China and Vietnam, monitoring and analyzing those children\'s mental health, which covers emotional expression, interpersonal interaction and personality defects. Methods: By the ways of field trip, questionnaires and etc.. Results: In fact, there are many factors affecting the mental health of left-behind children in cross-border ethnic groups, including social factors, family factors, school factors and so on. The results show that these left-behind children have different degrees of psychological problems. Conclusions: According to the results of the study, the author puts forward some measures of sports intervention. Firstly, help students know themselves well by exerting sports games. Secondly, promote students’ sense of belongings by developing team sports. Thirdly, help students foster the correct world outlook, views on life and values by popularizing the spirit of sports. Fourthly, enrich the students’ emotions by guiding parents to participate in sports. Through these effective measures, we try to keep the children left behind in the border of China and Vietnam physical and mental healthy. Experimental research on the influence of comprehensive intervention on cognition and behavior of pupils in physical education under the guidance of social ecology theory Jian Yang1,2, Cong Xiao1,2 1School of Physical Education and Health, East China Normal University, Shanghai, China; 2No. 1 Primary School Affiliated to Shanghai Normal University, Shanghai, China. Objectives: To explore the intervention effect of multi-faceted intervention on teenager physical exercise cognition and behavior from the perspective of social ecology theory to promote young people to actively participate in physical exercise. Methods: 173 young people from two schools in Wuxi City of Jiangsu Province were selected as the experimental subjects, including 85 experimental group students and 88 control group students, we tried to guide the experimental group to conduct a 14-weeks health education, exercise intervention and policy advocacy from the family, school, community and policy aspects. Results: 1) The number of participants with low level of participation in physical exercise after the experiment is reduced by nearly 23% compared with that before the experiment, which is mainly manifested in significant increase in time (P 0.01) and exercise frequency of participating physical exercise (P 0.01), but no significant change in exercise intensity (P 0.05). After the experiment, the value cognitive of exercise (P 0.01), emotional experience (P 0.05) and behavioral control (P 0.05) of the experimental subjects had a significant change, the behavioral attitude change (P 0.05) was not obvious. 2) Comprehensive intervention for Junior high school students to participate in physical activity time is better than primary students, both boys and girls played a significant effect comprehensive intervention for Junior high school students, primary school student frequency of exercise is obvious, the frequency of girls to exercise intervention is better than boys. 3) The effect of comprehensive intervention on the cognitive value of the exercise value of primary school students was significantly better than that of Junior high school students, and there was a significant intervention effect on both boys and girls. Comprehensive intervention effect of emotional experience for primary school students is obvious, the effect of intervention in junior high school students is not significant, and emotional experience intervention for boys is better than girls. Intervention on junior high school students’ exercise control was significantly better than that of primary school students; both boys and girls had significant effect and the intervention for boys is better than girls. Conclusions: The comprehensive intervention method can promote the cognition and behavior of physical exercise in primary and middle school students in a short period of time under the guidance of social ecology theory. In the future, we need to adopt a more comprehensive intervention method, develop a more rigorous intervention program, expand the sample size and prolong the intervention time to further explore the impact of comprehensive intervention on cognition and behavior of physical exercise from the perspective of social ecology and then come to more convincing conclusions. Acknowledgements: This work was supported by the Key Project of Shanghai School Physical Education Research in 2018 (Grants No. HJTY-2018-B03). A meta-analysis of effect of the attention bias modification on anxiety and depression Chih-Yun Kuo1, Wan-Lan Chen1, Ivo Stachiv2,3 1Department of Human Development and Psychology, Tzu-Chi University, Hualian City, Taiwan (R.O.C.); 2School of Science, Harbin Institute of Technology, Shenzhen School, Shenzhen, China; 3Institute of Physics, Czech Academy of Sciences, Prague, Czech Rep. Objectives: The attention bias has been proven playing a key role in the onset and maintenance of anxiety and depression. Attention bias modification (ABM) is widespread experimental paradigm that enables individuals to reduce selective attention to threat inducing by training the adaptive and maladaptive attention biases. Despite the fact that the ABM gained a considerable attention in past decade as either an experimental paradigm or a form of treatment; the quantitative understanding of ABM effects on anxiety and depression is still missing. Here authors examine whether the ABM can greatly reduces the anxiety and depression syndromes. Methods: We use standard meta-analytic procedure to investigate the effects of ABM on reduction of anxiety and depression syndromes. Particularly, by a systematic search of attention bias studies related to anxiety and depression published from 2000 to 2018, we identified 23 (1291 participants) ABM relevant articles. Then, we employ statistical analysis to examine effect of ABM on anxiety and depression. Results: Results show that ABM has a significant effect only on anxiety (only minor effect on depression was found) just when anxiety and depression are examined separately. Moreover, ABM for face stimuli poses essentially larger effect on anxiety syndromes than obtained by word stimuli. Conclusions: We have examined the effect of ABM on anxiety and depression syndromes. It has been shown that the ABM holds promises for treatment of anxiety. Age and gender did not moderate the effect of ABM treatment. Acknowledgements: This work was supported by the HIT-Shenzhen start-up project. A study of the effects of different pedaling frequencies on heart rate and muscle fatigue under the same load YuChe Huang1, SyuanLan Shih1, Xiaohong Yu2 1Department of Industrial Design, Chaoyang University of Technology, Taichung, Taiwan (ROC); 2School of Economics and Management, Shandong Yingcai University, Jinan, China Objectives: Most obese people use bicycle exercise to improve their physiological metabolism, further reducing body weight and fat, but also cause injury to the shoulder, neck, lower back, knee or wrist joints due to wrong training, and the sacral friction syndrome It is the most common sports injury. This study hopes to provide a reference for lower limb training in obese people in the future. Methods: The subjects were 20 subjects aged 20-35years old. Through the heart rate measurement system, wireless muscle power measurement system, lower limb pedaling force sensing device, etc., the experiment was completed according to different pedaling speed frequencies. Different pedaling frequencies, so each experiment is separated by 2weeks, avoiding muscle fatigue or training effect affecting the experiment, while the pedaling frequency is randomly selected. The statistical analysis of the data is using SPSS15.0 to analyze the variables according to Pearson correlation analysis. Compare and find the fatigue index by linear regression analysis. Results: The results of this study show that the average heart rate at 60 RPM is greater than the pedaling frequency of 40 RPM and greater than the pedaling frequency of 20 RPM. Therefore, it can be inferred that under the same lower limb load, the faster pedaling frequency will result in a higher heart rate load. If the data were analyzed by Pearson correlation method, it can be found that there is no positive correlation between heart rate increase and muscle fatigue, that is, there is a negative correlation between MDF of the tibialis anterior muscle and MDF of the biceps femoris. Conclusions: From the results of the study, it can be seen that under the same lower limb resistance, there is significant fatigue (specific fatigue index) for the specific muscle group under different cadence frequencies, and the heart rate curve and MDF are analyzed, and heart rate changes and muscle fatigue are found. Negative correlation (the longer the riding time, the heart rate increases but the MDF decreases), it can be seen that the fatigue of the body muscles is not positively correlated with the changes in the physiological heart rate. Therefore, in the future, for obese people who want to reduce body weight and increase physiological metabolism through lower extremity exercise, when feeling the heart rate increases, for muscle fatigue or physical fatigue, it is possible to extend the exercise time, but in the future, it is necessary to The physiological response, such as maximal heartbeat, is further studied to reduce the burden on the heart and the resulting risks. The impact of mind-body exercise (Sun-style Tai Chi Chuan) on college students’ anxiety and stress Mengmeng Liu, Yun Cao Dept. Of Physical Education, Zunyi Normal College, Zunyi, China Objectives: Recently, there has been an escalating trend in Chinese adolescents’ psychological problems, which include emotional problems such as anxiety, stress and so on. As predicted by World Health Organization, by 2020 there will be an increase of 50% in the number of adolescents with psychological problems worldwide, which ranks in the top five causes of adolescents’ diseases, disabilities and deaths. Therefore, adolescents’ psychological problems demand immediate attention. As a mind-body exercise, sun-style tai chi chuan originated in China, and is characterized by slow gentle motions including physical and respiratory movements. Extensive research indicates that Sun-style tai chi chuan is conducive to the mind and body. In addition, if people do tai chi exercise often, it is also beneficial for people\'s health as well. This paper aims to investigate the impact of Sun-style tai chi chuan on college students’ anxiety, stress as well as explore the relationship between psychological resilience and anxiety and stress respectively. Methods: 50 participants were selected from Zunyi Normal College, Zunyi Medical College. Randomly divided the 50 participants into Tai chi Group (Sun-style Tai chi chuan) and the Control Group (Health Education) and signed informed consents. The Tai chi Group practices three times a week for 12weeks (1hour each time by aerobic exercise) after mastering the basics of Sun-style Tai chi chuan and both two groups could not attend other exercise but Sun-style tai chi chuan after the intervention. We conducted pre- and post- tests of the two groups by means of the Self-Rating Anxiety Scale (SAS), Stress Scale for College Students (SSCS) and the Connor-Davidson Resilience Scale (CD-RISC). Results: 12-week Sun-style Tai chi chuan can alleviate college students’ stress level, but has no significant difference in easing anxiety. In addition, psychological resilience is negatively correlated to anxiety and stress respectively. Conclusions: The results of the Sun-style Tai chi chuan intervention indicate that it is beneficial to mitigate college students’ stress, but not significantly in anxiety, this may be partially attributed to the short experimental duration. The experimental results would be more significant if the duration were 24 or 48weeks. In addition, the result provides certain evidence that the intervention of other exercise for people suffering from mental problem would be extremely beneficial. The effects of open and closed exercise manners on psychological stress for college students Feixing Duan, Qing Wu Dept. of Physical Education, Wuhan University of Technology, Wuhan, China Objectives: As one kind of mental problem, huge stress can lead one falls into a sub-health state. In recent years, an increasingly number of abnormal behaviors is in play. College student suicide and hurt themself have becoming a serious issue. According to the investigation, the reason why they suicide or self-injury is because they feel a lot of stress. Apparently, huge stress has put college students in a dangerous situation, threatening their study, life and so on. There are some researchers showed that physical exercise is good for individuals’ mental health. The aim of our study is to use open and closed exercise for college students, finding out the differences between open exercise and closed exercise and figuring out a more excellent exercise option for stressful college students to ease their stress level and improving their quality of life. Methods: 90 participants were selected from Wuhan University of Technology. We used a randomized control trail design, with participants assigned in a 1:1 ratio to three groups, control group with usual life style (n=30), closed group with Baduanjin and usual life style (n=30), open group with basketball and usual life style (n=30). In the case of mastering basketball and Baduanjin exercise, the closed and open group received a 12weeks, 4 times for par week and at least one hour each time exercise training, control group kept their usual life style. Stress Scale for College Students (SSCS) are used to test the subjects before and after the intervention. We used SPSS23.0 to analysis the test data, finding out the differences between closed and open exercise for college students in easing their stress level. Results: The results showed that both health qigong and basketball could ease college students’ stress level. Comparison the differences between health qigong (Baduanjin) and basketball, health qigong is less effective in easing stress level than basketball. Conclusions: We could conclude that both open and closed exercise could lower college students’ stress level, but open exercise is more beneficial than closed exercise in easing stress. In the same kind of mental study, our study research could give a reference. Acknowledgements: This work was supported by The Fundamental Research Funds for the Central Universities (WUT: 2017VI023). Relationship between adolescent sports idols identification and social adaptation: the mediation effect of self-identity Tian Huang, Fen Qiu, Fei Du Wuhan University of Technology, Wuhan, China Objective: Sports stars are increasingly becoming an important idol choice for young people. Blind worship in the process of interaction between teenagers and social environment will cause certain crises to individual psychological development. The failure of establishing self-identity will directly affect the personality improvement and social adaptation of young people. In order to play the sports star idol positive guidance role and improve the social adaptation ability of young people, this paper will discuss the relationship between adolescent sports idols identification and social adaptation. Methods: In this study, the sports idol identity scale (Ma Rui, 2015), the identity crisis scale (Zhang Qingfang, 2002) and the Social adaptation scale (Huang Huahua, 2014) were used to measure the sports idols identity, self-identity and social adaptation of 729 high school students in Wuhan University. Three different surveys were conducted between one week, the first one was adolescent sports idols identification the second was Self-identity, and the third was Social adaptation to avoid interaction. And 20 high school students were interviewed, the data were sorted and analyzed by mathematical statistics. Results: The results showed that the degree of low involvement of sports idols was positively correlated with the scores of social adaptation, and the degree of highly involved sports idols was negatively correlated with the scores of social adaptation. The identity of essence in sports idols is positively correlated with interpersonal adaptation, emotional adaptation, and self-adaptation, and there is a significant impact (β=0.29, 0.15, 0.33, P value are less than 0.01). Sports identity is positively correlated with interpersonal adaptation and self-adaptation, and there is a significant impact (β=0.17, 0.14, P value are 0.01). There is a negative correlation between the identity of essence and self-identity in sports idol of adolescent,and there is a significant impact (β=−0.09, P value are 0.05); while sports identity and appearance identity are negatively correlated with self-identity, and there is no significant impact (P value are 0.05).Self-identity of adolescents is positively correlated with interpersonal adaptation, emotional adaption, self-adaptation, and there is a significant impact between self-adaptation and self-identity (β=0.23, 0.18, 0.25 P value are 0.05). Conclusion: adolescent sports idol identity has direct influence on social adaptation, and self-identity plays an indirect role in adolescent sports idol identity and social adaptation. That is self-identity plays the mediating effect in adolescent sports idols identification and social adaptation. Keywords: Self identity; Sports Idol identity; Social adaptation; adolescents. 1Dept. Computer Software Engineering, Changshin University, Changwon-Si, South Korea; 2The Bradley Department of Electrical and Computer Engineering, Virginia Tech, VA, USA Objectives: Most people spend most of their time a day working indoors (e.g. factory, office building, etc.). Therefore, it is important to develop an efficient indoor location estimation algorithm to support various location-based applications. However, there is a need for an alternative to using outdoor space because users have to follow an indoor floor plan to move indoors. In this paper, we propose location estimation algorithm based on Bayesian filtering for indoor location estimation using noisy RFID raw data. Methods: We deployed an event-based data collector at the beginning of the system to identify the RFID raw reading. The data collector module efficiently stores RFID raw readings for subsequent query processing operations. The data collector module collects raw readings in time units of one second to keep them organized. This is because Bayesian filtering does not require high observation frequencies. We made two models, the indoor moving graph model and the fiducial pointing model to track the object locations in indoor environments using an RFID-based system. The basic approach to kNN query or an all-range estimation is to calculate the location probability distribution of all objects in the current indoor configuration. Results: We evaluated the scalability of the proposed algorithm by changing the number of moving objects from 100 to 1000. The average hit rate of the kNN queries decreased in all methods, but the KL variance of the three methods is relatively stable. The decrease in kNN hit rate is due to the increase in density of the object. We tested the effect of the activation range of reader with varying the range from 30cm to 300cm. As the activation range increased, the uncertainty areas not covered by all the readers were essentially reduced, resulting in a 10% improvement in the performance of the three methods. Conclusions: We proposed a location estimation algorithm based on Bayesian filtering in order to calculate accurate indoor position with noisy RFID reading. Experiments using data generated by actual parameters have shown that the accuracy of the proposed algorithm is superior to the conventional symbol model based method. Acknowledgements: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NRF-2018R1C1B5046282). Research on the effect of physical exercise on improving adolescents’ mental health and interpersonal relationship Zhonghao Li, Fen Qiu Wuhan University of Technology, Wuhan, Hubei, China Background: Adolescence is a critical period of psychological and behavioral problems in the process of individual growth and development, and the level of mental health in this period affects its lifelong development. The pressure of their study and life is constantly increasing, which directly leads to the tension of interpersonal relations. And produce certain bad emotions, such as jealousy, dissatisfaction, hostility and so on. Therefore, the impact on the mental health of adolescents is a research topic worthy of further exploration. Methods: In this study, 490 investigation reports were analyzed by using \"physical exercise rating scale”, \"Adolescent Self-rated Life event scale”, \"Comprehensive Diagnostic scale of interpersonal relationship” and \"College students Mental Health scale”. The data were analyzed by systematic statistics and variance analysis to explore the current situation of adolescent physical exercise and the general characteristics and changes of life stress resistance, mental health and interpersonal relationship. Results: The results of this study showed that the amount of physical activity was positively correlated with the stress tolerance of adolescents (P 0.05), which was positively correlated with mental health (P 0.05). There was no significant correlation with interpersonal relationship (P 0.05). Conclusions: Physical exercise is very important to improve adolescent\'s ability to resist stress, mental health and interpersonal relationship. Through physical exercise, young people can not only improve the overall body function, improve the ability of mental activity; Secondly, through physical exercise, to provide more opportunities for young people to communicate with each other, to promote the improvement of young people\'s social adaptability, so that young people can correctly face all kinds of setbacks and difficulties in life. Form an independent personality; finally, through physical exercise, but also to further promote the psychological health of young people, so that young people have a healthy mentality to face life. Objectives: The technique of extracting human key points based on partial affinity field (PAF) is providing a new real-time count detection method basis for exercise training especially the strength training based on semi-squats or squats in the video. It can effectively avoid the discomfort with traditional wearable sensors and reach higher accuracy rate, in addition also used for multi-person exercise behavior counting at the same time. Methods: This method firstly proposes a motion vector field model based on the key points of human bodies (including 1-Head; 2-Neck; 3-Hip; 4-Left Knee; 5-Reft Knee; Note that 3, 4, 5 compose a triangle that shows human body′s movement stability in some extent), as shown Fig.1. The key points that are significantly moving with the individuals of human body are then associated in the video by learning in the PAF. This step is more important for multi-person training scenarios, and the result of detection is shown Fig.1(a). Further, the method constructs a motion vector field of a body key point of each of the moving individual in the video using the motion feature sequence of the continuous video frames. Finally, the motion vector field is used for count the squats behavior within a certain period. Results: The effectiveness and feasibility of our method is proved by typical semi-squats and squats exercise videos, which choosing from different angles such as the front and the side view for experiments. The experiment results are shown in Fig 2. It is easy to discover that exists squat behavior characteristics of periodic motion which an approximately regular squiggle goes up and down in one dimensional continuous time axis from the trajectory of vector sequence of movement. Finally, according to the motion curves the amount of squat exercises is detected correctly. The results further test that it is also suitable to detect of the videos from side view, so the method has good adaptability. Figure1: The detected model of key points of human bodies (a) Multi-person training scenario in front view (b) single-person training scenario in side view Conclusions: Our approach provides the new detection strategy using computer vision technology, and solves the limitations of traditional count sensors, particularly in the real-world multi-person exercises scenarios. Experimental results show that proposed method is a positive exploration on the measures of exercises amount, furthermore reasonable and feasible. 1Instrumental Analysis Center of IMUN, Inner Mongolia University for the Nationalities, Tongliao, China; 2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: In the course of neural network model training, deep learning technique often leads to the problem of fitting. Overfitting means that the model fits the data of training set very well, but it does not fit the data set that it has not learned well, and its generalization ability is weak. That is to say, the effect of the sample is very good, and it is not good to extend to the more general sample. Methods: Aiming at the over-fitting problem in neural network model, a convolutional neural network is proposed, which combines maximum pooling of differentiated overlapping regions with dropout technique. The maximum pooling of differentiated overlapped regions can alleviate the overfitting problem more effectively. Results: The method proposed in this paper is tested on the CIFAR-10 dataset. The dataset contains 60000 color images, which are 32*32, divided into 10 classes and 6000 diagrams for each class. There are 50000 pieces for training, which constitutes 5 training batches, each batch of 10000 images. Another 10000 is used for testing, and they form a single batch. For the two models of convolution neural network based on average value and maximum value non-overlapping pooling, the maximum pooling is better than the average pooling, but both of them have over-fitting problems. The convolutional neural network based on the maximum overlap pooling reduces the overfitting problem to some extent. This paper presents a convolution neural network based on the maximal pooling and dropout technology of the overlapped overlapping region, and through the experimental results we can see that the accuracy rate of the training set is much lower than the test set, and its accuracy is higher. Conclusions: Compared with the over-fitting problem of traditional convolution neural network, the convolution neural network proposed in this paper improves the image classification performance of the model for image classification, and achieves better classification results on CIFAR-10 data set. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). Use of solar energy and ITC to reduce the carbon footprint in irrigation systems. The case study on a semiarid zone (Pliego) in the southeast of Spain Chazarra-Zapata, Jesús Pedro1, Egea Pérez, Ramón2, Melia-Navarro, Amparo1, Ruíz- Canales, Antonio1, López-Peñalver, Francisco José3 1Engineering Department, Miguel Hernández University of Elche, Orihuela (Alicante), Spain; 2Dr. Civil Engineering, EMUASA, Municipal Water and Sanitation Company, Murcia, Spain; 3Master Engineering Student, Superior Polytechnic School, Alicante University, Spain. Objectives: Due to the geographical characteristics and its irradiation index, the South of the European Union (EU) delimited by the Mediterranean Sea, is a perfect scenario for the employment of renewal energies. One of the applied fields where these energies can be used is the Agriculture. Moreover, Spain is the country of the EU with the highest number of annual sunshine hours. This allows obtaining a biggest efficiency in solar panel use. The work will be finished with a comparison of economic saving and return on investment after the implementation of photovoltaic energy in our study area. Additionally the reduction of the carbon footprint in this cases study according to the last Unit Nations Conference about Global Change will be analyzed. Methods: Firstly we need to know climatology, thermal regime, agroclimatic traits, and precipitation regime, for it has been studied the water needs and available resources and the economical expenses of the electric consumptions of different sources. Secondly, the Co2 emissions generated with the existing operation have been valued. Third, the possible measures to be taken to reduce emissions are studied As a result of these data it is obtained that in our case the best alternative is the use of solar energy through the design of a photovoltaic system controlled by sensors and ICT that allows us to optimize this energy applied to the governance of water and enhancing the use of reclaimed water for agricultural production of high quality fruit. It should be noted that in semi-arid areas of the Mediterranean we should not only think of fruit / agricultural plantations as primary means of production, but also as an ecological method of protection against climate change, particularly against desertification. In summary, this study seeks to collaborate to meet the three objectives of European policy within the Framework on Climate and Energy by 2030. Conclusions: the application of these measures means an annual reduction of 782.5 tons of CO2 eqv not discharged into the atmosphere. These methods of primary production, such as agriculture, must be integrated into sustainable technological development and their objective is to mitigate climate change and improve the uses of reclaimed water applied to agriculture. Acknowledgements: Our gratitude the Region of Murcia, the participating Water User\'s Association of Pliego, the Company Los Serranos-Group Eiffage and especially to Miguel Angel del Amor Saavedra. I am grateful for the ALICE Project \"Accelerate innovation in urban wastewater management for Climate change” - Proposal Number: 734560, financed by European Commission, Grant of Horizon2020 under the specific programme Marie Sklodowska Curie Actions -RISE-. Production of fructo-oligosaccharides from levan treated with compressed hot water fluids Andres Abea1, Tetsuya Ushiyama1, Ebru Toksoy Öner2, Naoto Shimizu3,4 1Graduate School of Agriculture, Hokkaido University, Hokkaido, 060-8589, Japan; 2IBSB, Department of Bioengineering, Marmara University, Istanbul, Turkey; 3Research Faculty of Agriculture, Hokkaido University, Hokkaido, 060-8589, Japan; 4Field Science Center for Northern Biosphere, Hokkaido University, 060-0811, Japan Objectives: Fructooligosaccharides (FOSs) obtained from the hydrolysis of fructans can be used as additives in the food industry, for example as sweeteners or dietary fiber. Given its capacity to improve gut absorption of calcium and magnesium, to prevent urinal infections (by promoting proliferation of lactobacilli), to reduce the risk of colon cancer, to enhance lipid metabolism, and to attenuate the development of tooth decay, FOS can be considered a functional food ingredient. Commonly employed methods of extraction of oligosaccharides are hydrolysis of fructans or synthesis from disaccharide substrates by enzymatic and chemical treatment. Compressed hot water can be used as an alternative process for more sustainable hydrolysis of polysaccharides than processes that use acids or enzymes. Methods: Levans from Erwinia herbicola (EH) and Halomonas smyrnensis (HS) were characterized using scanning electron microscopy and light scattering techniques and then hydrolyzed using compressed hot water at each of four temperatures (120, 140, 160, and 180°C). The hydrolysates were analyzed using high-performance liquid chromatography (HPLC) and electrospray ionization-mass spectrometry. Results: Levan HS showed a crystalline morphology, whereas levan EH showed an aggregated structure. Both levans had molar masses on the order of 106g/mol, but levan EH had a smaller radius of gyration, hydrodynamic radius, and intrinsic viscosity. The hydrolysis results showed that levan EH started hydrolyzing into FOSs at approximately 120°C, while levan HS required a temperature of at least 160°C, possibly due to differences in the degree of branching of the two levans. Both samples were degraded to fructose when treated at 180°C. Conclusions: In the treatment of fructans with compressed hot water fluids, the molecular conformation of the substrate must be considered when deciding the reaction conditions needed to obtain the desired products. Acknowledgements: The fiber DLS and the SEC-MALS analyses were performed by the instrumental analysis services of the Global Facility Center of Hokkaido University. Evolutionary nonnegative matrix factorization algorithms for community detection in dynamic networks Yu Pan1,2, Zhixiong Xu1, Luogeng Tian3 1College of Command and Control Engineering, Army Engineering University, Nanjing, China; 2Communications Support Unit, Shenyang, China; 3Xi\'an Research Institute of High-Tech, Xi\'an, China Objectives: Complex networks are employed in many fields to represent various kinds of complex systems, such as biology, medicine and computer. Dynamic community detection in complex networks can analyze the evolutionary mechanism of complex networks, predict the evolutionary trend of the community, analyze the user behavior and then provide support for information dissemination and user behavior prediction. For instance, in disease networks, cancer cell migration leads to metastasis, which is critical for cancer progression and therapy. Many approaches have been proposed to detect the communities in static networks network. However, most of the proposed approaches unable effectively detect the communities in dynamic networks. Based on the idea of evolutionary clustering, we propose a dynamic community detection algorithm based on Nonnegative Matrix Factorization (NMF) Evolutional and the algorithm can accurately and quickly detect meaningful communities in dynamic networks. Methods: We propose a dynamic community detection algorithm based on Evolutionary NMF. The algorithm adjusts the network topology of the current moment according to the community detection at the historical moment to obtain a smoother adjacency matrix. Firstly, we propose a network topology adjustment strategy: When the objective function is optimized, the community indicator matrix of the current moment is initialized by the community detection of the historical moment, which reduces the number of iterations and improves the efficiency of the community detection. Results: The proposed algorithm is compared to two existing dynamic algorithms FacetNet (A Framework for Analyzing Communities and Evolutions in dynamic Networks) and sE-NMF (Semi-supervised Evolutionary NMF) on five synthetic datasets. LFR model is used to generate dynamic networks with constant, reduced and increased community numbers. Experiments show that the proposed algorithm has a significant improvement in the Accuracy and NMI (Normalized Mutual Information) indicators compared to the other two algorithms. Among them, in the third moment of synthetic dataset 2, our algorithm improves the value of accuracy by up to 19.7%, 21.3% and value of NMI by up to 17.2%,19.6% compared with the other two algorithms. Conclusions: We have recently witnessed the emergence of dynamic networks in the real world, providing a huge opportunity to identify evolving patterns. For dynamic network, we propose an effective dynamic community detection algorithm based on evolutionary NMF. The proposed dynamic community detection algorithm can be accurately and quickly mine the community structure at each moment, which is suitable for dynamically changing network scenarios. 1Instrumental Analysis Center of IMUN, Inner Mongolia University for the Nationalities, Tongliao, China2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives The multi product and small batch production for customer ordering is the leading production mode at present. Accordingly, manufacturing enterprises are developing towards lean production and agile manufacturing. Under this kind of production environment, how to arrange the production plan and how to schedule has become the key issue for the smooth and effective production of enterprises. Effective scheduling methods and optimization techniques play an important role in improving production efficiency and reducing production costs for manufacturing enterprises. Methods: Based on the research status of workshop scheduling problems at home and abroad, this paper takes the actual situation of the current operation workshop as the research object, and conducts in-depth systematic research on the production scheduling problem of the operation workshop. Firstly, aiming at the limitation of the classical shop scheduling problem, the optimal scheduling problem with flexible paths in advanced manufacturing workshops under FMS environment is studied. Considering the multi-objective characteristics of job-shop scheduling problem and the requirement of modern production just-in-time system, a bi-objective mathematical optimization model for flexible job-shop scheduling problem is established, and a modeling example is given with the Kailing project. Secondly, a flexible shop scheduling model based on genetic algorithm is proposed. One of the most important topics in constructing genetic algorithms for job shop scheduling is to design appropriate solutions and genetic operators based on specific problems, so that all chromosomes generated in the initial or evolutionary process will produce feasible scheduling. Results: The parallel GA algorithm, which satisfies the constraints, can complete the computation of the scheduling algorithm in less than 1second. For large-scale problems, the control parameters such as the number of operations, population size and so on can still obtain satisfactory results, but the computational time is longer than the previous example. But for the actual shop dynamic scheduling problem, the calculation time of 1second and 10seconds or even tens of seconds is not much different, and does not affect the normal operation of the job shop. Therefore, parallel GA algorithm can meet the requirements of dynamic scheduling. Conclusions: On the basis of theoretical research, this paper establishes the dynamic scheduling model of production workshop, designs and implements the genetic algorithm for the scheduling problem, and analyzes and designs the dynamic scheduling method. These research results lay a foundation for further research and development of workshop scheduling theory and practical scheduling software. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). 1Southwest China Institute of Electronic Technology, Chengdu, China; 2College of Computer, Sichuan University, Chengdu, China Objectives: With the explosion of trading volume of various digital currencies represented by Bitcoin and Ethereum in daily life, the security risks and transaction risks of these digital currency applications based on blockchain technology are also growing. These security issues focus on the two most important participants in the blockchain architecture: both trading users and miners of block computing. It mainly attacks on two levels of blockchain transaction messages and block synchronization, such as garbage transactions based on block transactions, dust attacks, etc., based on long-chain attacks and block fork attacks. These security issues seriously threaten the correctness and integrity of blockchain security operations and blockchain data, causing huge losses to users who use digital currency for transactions and miners who participate in blockchain accounting, and have serious impacts on the promotion of blockchain technology and the development of digital currency. Therefore, it is extremely urgent to improve the security of blockchain transactions and data blocks. Methods: Artificial immune system is one of the key technologies to solve the problem of network security because of its high immune protection, immune tolerance, immune surveillance and other security protection, security detection, self-learning and memory functions. At the same time, as well as the decentralization of the blockchain, and the widespread distribution of the miners involved in the blockchain, this has an amazing similarity with the distribution, self-adaptability and diversity of artificial immunity. In order to solve the security problem of blockchain in transaction and block, this paper proposes a dual-layer blockchain safe immune detection model based on artificial immune theory. The model identifies two levels of antigens, antibodies, detectors, affinity algorithms, etc. by establishing artificial immune models at the two levels of trading and block. The dual-layered secure immune model enables artificial immune protection of blockchains in transaction distribution and verification, block synchronization and so on. Results: After the establishment of the dual-layer immune model, the related parameters such as affinity, population size and memory are optimized by learning the historical data occurring in the blockchain. By adding the dual-layer immune mechanism to the open source bitcoin code, the RPC method is used to send a large amount of junk transactions, and the transaction layer\'s immune mechanism can identify the attack; by sending long-chain attacks between multiple miners, the block-level\'s immune mechanisms is triggered and automatically isolates malicious nodes that send out malicious transactions and block attacks. Conclusions: According to the comparative experiments, the dual-layer immune security model can effectively identify the transactions and blocks attacks, so as to improve the transaction security and blockchain security and avoid the impact of security attacks on blockchain. Acknowledgements: This work is supported by the Fundamental Research Funds for the Central Universities (No. YJ201727). Robot obstacle avoidance and navigation control research based on multi-sensor information fusion Xiaojing Fan1, Mingyang Jiang2, Zhili Pei2 1College of Mechanical Engineering, Inner Mongolia University for the Nationalities, Tongliao, China; 2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: This paper presents a multi-sensor information fusion algorithm based on T-S (Takagi-Sugeno) fuzzy neural network. The algorithm combines the advantages of neural network and fuzzy logic, has the ability of fuzzy reasoning to process uncertain information and the ability to learn according to sample data, and can better realize the autonomous obstacle avoidance and navigation of robot in unknown environment. Methods: Firstly, the sensor system of mobile robot is constructed, the information of obstacle in unknown environment is provided by ultrasonic sensor and infrared sensor, and the positioning system composed of electronic compass and GPS provides location and target information of robot. Then, the fuzzy logic and T-S fuzzy neural network are applied to the obstacle avoidance and navigation control of the robot, and the two control methods are simulated and compared with MATLAB. In the fuzzy logic information fusion algorithm, the language value of each variable and its membership function are calculated according to the function of each sensor in obstacle avoidance and navigation system of mobile robot and the actual situation of measurement information. Based on the navigation control experience of navigation experts and operators, the fuzzy control rules of navigation for mobile robots are established. Finally, a fuzzy inference system for obstacle avoidance and navigation control of mobile robot is established. In the T-S fuzzy neural network information fusion algorithm, the simplified structure is adopted. The network of robot obstacle avoidance and navigation control was trained with 100 sets of data as training samples. Results: T-S Fuzzy neural network integrates the advantages of fuzzy logic and artificial neural network, and improves the ability of the system to deal with the environment information collected by sensors. This method can produce real-time and accurate control decision. The experiments show that the T-S fuzzy neural network information fusion algorithm is better than the fuzzy logic information fusion algorithm in the obstacle avoidance and navigation control of mobile robot. Conclusions: Multi-sensor Information fusion algorithm based on fuzzy logic and T-S fuzzy neural network is effective in obstacle avoidance and navigation control of mobile robot. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). Visualization of the image geometric transformation under the illumination variation based on the product manifold Tianci Liu1,2,3,4,5, Zelin Shi1,2,4,5, Yunpeng Liu1,2,4,5, Chenxi Li1,2,3,4,5 1Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China; 2Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, China; 3University of Chinese Academy of Sciences, Beijing 100049, China4Key Laboratory of Opto-Electronic Information Processing, Chinese Academy of Sciences, Shenyang 110016, China5The Key Laboratory of Image Understanding and Computer Vision, Liaoning Province, Shenyang 110016, China Objectives: The geometric transformations in images are abstract and difficult to comprehend and grasp. In the natural conditions, the geometric transformations of objects in images are always under illumination variation, which leads to more model complexity for visualization. Thus, an appropriate method should be presented to model this transformation and visualize its changing process for comprehension. This study aims to provide a novel method to solve this issue. Methods: We modeled our framework based on the product manifold. The geometric transformations were computed based on the SL(3) group and the illumination changes can be modeled as the affine transformation group GA(1). However, the distances of projective transformations with the illumination changes cannot be measured directly because of the structural difference of these two manifolds. To visualize this combined geometry, we modeled our framework on the product manifold (i.e., SL(3) × GA(1)) and designed a new metric which can measure the distance of points on the product manifold. Based on this new metric, we embedded these points on the original product manifold into the three-dimensional space based on the multidimensional scaling (MDS), which preserves the original distance on the original space. Results: Based on our new designed metric and approach, the visualization results of the various manifold exhibited as following: 1. The visualization results of the illumination changes based on the affine transformation group GA(1) are exclusively shown in Figure1. 2. The visualization results of the projection transformation based on SL(3) group are exclusively shown in Figure2. 3. The visualization results of the combined model based on the product manifold are shown in Figure3. 4. The distribution histograms of distances among the actual points on the product manifold and computed coordinate-points in low-dimensional embedding space are shown in Figure4, respectively. Conclusions: We present a new metric and novel method to visualize the projection transformations in images based on the product manifold. Three various manifolds are visualized to show the changing process of different transformations. From the results, the proposed framework well approximates the projection transformations under the illumination changes and preserves its structure desirably. Acknowledgements: This work was supported by the Innovation Fund of Chinese Academy of Sciences (Grant No. Y8K4160401). 1Chemistry and Chemical Engineering College, Inner Mongolia University for the Nationalities, Tongliao, China2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: In the era of large data, text data accounts for a considerable proportion. As an effective way to manage and organize text data, classification has gradually become a focus of attention. Aiming at the disadvantage that the classification speed and accuracy can not be considered simultaneously, a text classification method based on improved DNN-KNN is proposed in this paper. Methods: Deep neural network model (DNN) is used to reduce the dimension of text data. Restricted Boltzmann machine is used to pre-process DNN, and conjugate gradient is used to fine-tune DNN. Data after dimensionality reduction is used as input to improve KNN model, and classification is completed. The main disadvantage of KNN is that it needs to calculate the distance between the text to be classified and all the training text. The time complexity is directly proportional to the number and dimension of the training text. It is expensive to calculate the similarity with a large number of training text. In this paper, an improved KNN classification algorithm (RS-KNN) based on rough set is proposed. On the basis of the traditional KNN algorithm, in order to solve the problem of large computational cost due to large sample set, the relevant theory of rough set is used to define each sample. The upper and lower approximation of the space reduces the sample search space, reduces the amount of calculation during classification, and improves the classification speed of the system. The improved algorithm mainly calculates the upper and lower approximate spatial radii ru and rl of each subset of the training text set according to the vector space model of the training text, and calculates the class center distance of the text to be classified and each text subset to determine which category it belongs to. . In order to verify the validity and correctness of the algorithm. Results: The experimental data set is the Reuters21578 corpus. In the test, 10 of them were selected. In the RS-KNN algorithm, the larger the lower approximation space, the more the number of test texts falling into the classification calculation process, and the smaller the calculation amount of the text similarity, if the lower approximation space is smaller, and falls into the boundary. The more the number of test texts in the paper, the more the text similarity calculation will be close to the traditional KNN algorithm. Since the search space of the text to be classified is reduced, the classification calculation overhead is reduced. In the case of a small amount of data, the classification speed of the improved algorithm is not obvious, but as the amount of data increases, the classification speed is significantly improved. Conclusions: The experimental results show that the proposed text classification method based on improved DNN-KNN has a significant improvement in classification speed and classification accuracy. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region(2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm(MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine(MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). Research on path planning of mobile robot based on genetic algorithm in dynamic environment Xiaojing Fan1, Mingyang Jiang2, Zhili Pei2 1College of Mechanical Engineering, Inner Mongolia University for the Nationalities, Tongliao, China2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: Aiming at the problem of robot path planning in dynamic environment, this paper presents a method based on genetic algorithm. A new operator and adaptive adjustment method are added to solve the problem that the evolutionary process can not reach the target point because of falling into local minima. This can solve the problem of path planning of mobile robot in dynamic environment. Methods: In order to solve the problem of robot path planning, in the design of genetic operators, smoothing, inserting and removing operators are added to complement the deficiencies of basic operators, and the genetic algorithm is optimized by the adaptive adjustment of mutation rate and crossover rate. The occurrence probability of crossover operator, mutation operator and smoothing operator is set, and the value of safety margin and deflection angle are set in the calculation of fitness function. By adding the new operators and adaptive adjustment method, the algorithm can be improved to solve the problem that the evolutionary process can not reach the target point because of falling into local minima. The experiment was conducted in 3 different environments with varying degrees of complexity. Results: In three kinds of complex environment, the search algorithm proposed in this paper has successfully found the approximate optimal feasible path, which shows that the proposed algorithm can adapt to all kinds of complex environments independently, with strong searching ability, fast convergence speed and stable performance. Compared with other methods, the path planning method based on genetic algorithm is superior to other algorithms in the same environment in terms of search time, path length and smoothness. In the experiment, the individual in the running population is evolving. The individual\'s fitness value is also constantly changing, and when the optimal path is found, it will stop evolving and exit the cycle. Conclusions: This paper studies the path planning method of mobile robot based on genetic algorithm in dynamic environment. By using smoothing operator, inserting operator, deletion operator and adaptive method, the convergence speed of path planning for complex environment is improved, and the local optimal solution can be avoided. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). Jiangang Ku1, Xianyang Liu1, Adrian R. Muxworthy2, Zixu Miao1, Miguel Valdez2, Huihuang Chen3, Rongdong Deng1, Weiran Zuo1, Wanzhong Yin1 1School of Zijin Mines, Fuzhou University, Fuzhou, China; 2Department of Earth Science and Engineering, Imperial College London, London, UK; 3School of Chemical Engineering, the University of Queensland, Brisbane, Australia Background: Geometrical shapes of magnetic mineral particles greatly affect the particle magnetic induction field (MIF), which is very important for modeling the processing of magnetic separation. However, the mechanism underpinning the formation of MIF is still poorly understood, not to mention an effective MIF model of real irregular-shaped magnetite particles. Methods: In this study, five geometric solids for regular-shaped magnetic particles were established, and the MIF of the particles was analyzed using finite element method (FEM). A new method is proposed to calculate the MIF of irregular magnetic particles by studying their magnetic flux density (MFD). Results: Sphericity is a key factor affecting the MIF of a regular-shaped particle. Moreover, the calculation value of the MIF using particle volume radius (r) is more accurate with the higher sphericity. Additionally, the relative position is another important factor affecting the accuracy of MIF calculation. The further the distance from the particle center, the more accurate the MIF. Furthermore, the relative difference between a particle MFD and the corresponding sphere MFD generally decreases with the increase in the sphericity and the distance from particle center. More specifically, when the distance is 4r, the average value of the relative difference drops to 0.40% and 0.01% with the sphericity of 0.81 and 0.91, respectively. In addition, the relative difference with the distance is very similar for all particles regardless of the variation of morphology, volume radius, and external magnetic field intensity. Conclusions: In this study, geometry shapes of magnetic mineral particles were analyzed and particles MFD were modeled based on FEM. The conclusion as follows: (1) MFD values drop rapidly with the increase of the distance before reaching a plateau (approximately zero) regardless of the sphericity. Furthermore, MFD values of particles with different sphericity vary greatly when the distance is less than 2r. When the distance is larger than 2r, MFD values are nearly equal except for tetrahedron along Z-axis, for which the MFD value is almost the same with other regular polyhedrons when the distance is larger than 3r. (2) The average value of relative differences between particles MFD and sphere MFD decreases rapidly at first, and then gradually declines to zero with the increase of the distance to the particle surface. In addition, it generally decreases with the sphericity. Thus, it is feasible and effective to use the MFD of a sphere to replace that of an irregular particle. (3) The changing trend for the relative difference between particles MFD from sphere MFD is the same for any particles regardless of the huge variation of volume radius or external magnetic field intensity. We suggest that it is feasible to use a sphere MIF instead of an irregular particle MIF to simulate the motion of massive magnetic particles. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grant No. 51674091), the Natural Science Foundation of Fujian Province of China (Grant No. 2017J01483), and National Training Program of Innovation and Entrepreneurship for Undergraduates (Grant No. 201810386029). Department of Information and Communications Engineering, Sejong University, Korea Objectives: In full-dimension (FD) beamforming scheme, beams are formed by two-dimension antenna array. Thus the system can switch the direction that the maximum power is transferred and also concentrate the radiated power more than conventional linear beamforming scheme to a targeted area. These properties contribute to power efficiency of the system and improve signal to interference plus noise power ratio (SINR) of received signals by mobile devices in cells. In this study, to verify the performance, the error rate of FD beamforming scheme is compared with existing analogue two-dimension beamforming scheme. Methods: To show the enhancement by full-dimension beamforming scheme, channel model that reflects three-dimension fading environment must be implemented. For the three-dimension channel model, the technical reports by the 3rd Generation Partnership Project (3GPP) are referred. The channel model proposed by 3GPP is the extended version of the previous two-dimension spatial channel model (SCM) of 3GPP. For planar antenna array, rectangular shape is assumed. The number of antennas for planar array is determined considering space for installation. The simulated array settings are 3 by 1 linear array, 9 by 1 linear array, and 3 by 3 planar array. These beamforming schemes are simulated on single-user multiple input multiple output (SU-MIMO) system by Matlab. The number of data streams is four and zero-forcing (ZF) detection scheme is used for demultiplexing the received signals at the receiver. Since the number of data streams is four, the total antenna numbers of simulated schemes are 12, 36, and 36. The number of antennas for the receiver is set to four and the total power for transmission is set to the same for all cases. Results: While diversity schemes increase the probability that high quality signals are received than received signals with non-diversity scheme, analogue beamforming schemes increase the received power more directly. Simulation results show the effect of increased received power. The curves align with the same slop. The scheme that shows the best performance is 3 by 3 planar and the next is 9 by 1 linear array. The gap between 3 by 3 array and 3 by 1 is 2dB and the gap between 3 by 3 and 9 by 1 is almost 1dB. 3 by 1 and 3 by 3 occupy the same space in the first dimension. Conclusions: In this study, FD beamforming scheme is compared with existing two-dimension beamforming scheme. FD beamforming scheme shows the best performance by using the same power. Also usage of the second dimension for antenna configuration increases spatial efficiency of system. Acknowledgements: This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program (IITP-2018-0-01423) supervised by the IITP (Institute for Information & communications Technology Promotion). The corresponding author of this paper is Hyoung-Kyu Song (songhk@sejong.ac.kr). Xiuxin Wang1, Shuaishuai Jiang1, Zhengbin Li2, Yuhao Jiang1, #, Zhangyong Li1, Wei Wang1 1Medical Electronics and Information Technology Engineering Research Center, Chongqing University of Posts and Telecommunications, Chongqing, China; 2State Key Laboratory of Advanced Optical Communication Systems and Networks, Beijing, China#The Author Contributed Equally to This Work and Should be Considered Co-first Authors Objectives: On the basis of self-built photoacoustic imaging system, experiments were conducted to study two-dimensional imaging with high spatial resolution. The novel photoacoustic sensor developed was a versatile tool for photoacoustic imaging. Methods: According to the response mechanism of microfiber Bragg grating Fabry-Perot interferometer, the microfiber photoacoustic sensor was fabricated by a suitable fiber for the characteristics of the device. Compared with the single mode fiber Bragg grating Fabry-Perot photoacoustic sensor, the detection capability of the microfiber Bragg grating Fabry-Perot photoacoustic sensor was analyzed. The dominant effect of mode refractive index under the action of large scale evanescent field was used to measure the photoacoustic signal, providing a guarantee for the realization of photoacoustic imaging. Results: Study the spectral characteristics of the microfiber Fabry-Perot interferometer and the influencing mechanism of the parameters such as fiber geometry size and grating reflectivity on the transmission spectrum. The cross sensitivity of temperature could be effectively removed by different response of the above parameters. The output voltage and power spectrum, frequency response spectrum and direction of photoacoustic sensor were comprehensively explained. The effective sensitivity enhancement method of microfiber Bragg grating Fabry-Perot photoacoustic sensor had been realized. Conclusions: The image of spatial light absorption distribution was reconstructed by experiments using optical sound transducer. High performance photoacoustic sensors with high sensitivity and high spatial resolution were realized. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grant No. 61605021); by Scientific and Technological Research Program of Chongqing Municipal Education Commission (Grant No. KJ1600441); by State Key Laboratory of Advanced Optical Communication Systems and Networks, China; in part by Doctoral Foundation of Chongqing University of Posts and Telecommunications (Grant No. A2015-26) and by Youth Foundation of Chongqing University of Posts and Telecommunications (Grant No. A2015-58). 1Guizhou Academy of Science, Guiyang, China; 2College of Computer Science and Technique, Huaibei Normal University, Huaibei, China Objectives: With the development of the research of food safety cloud platform and Internet technology, a kind of application in information sharing platform architecture is put forward. The different types of data are turned into a unified data model, and the heterogeneous data is integrated. Access control model based on multilevel shared groups is used in the platform. Methods: In order to establish a traceability management system for agricultural products, guide social forces to govern precisely, and use traceability data base to improve the efficiency of supervision information, the joint efforts of the government, enterprises, trade associations, social organizations, the media and consumers will manage the food safety platform cooperatively. The confidentiality levels of the subjects and objects are solved through the relationship between the shared groups and their relationships, which include supper of shared groups, lower of shared groups, the same shared groups and incomparable relationship. The integrity level of the subjects and objects are solved by zoning. Because the relation of shared groups can be changed, the model is flexible. First, A method of access control for hierarchical partitioned shared groups includes upper and lower levels of relationship and access policy in hierarchical partitioned shared groups. Second, the privileges are determined by the relation between the shared groups. The confidentiality level of the information is solved by zone of the shared group. The integrity level of the information is completed by the partitions within the group. Protecting privacy is carried out through the relation of shared groups. If the relation between the two shared groups is incomparable, then the security of them is incomparable. Finally, a case study is given to illustrate the effectiveness of the model. Results: The relation between the two shared groups shows the relative secrecy, which reflects the relative nature of the security. Changes in the relation reflect the dynamic nature of security. A complex problem is divided into some small problems, so that the problem is becoming simplification. Conclusions: Availability, confidentiality and integrity are harmonized in the food safety cloud platform. Acknowledgements: This work is supported by the National Key R&D Program of China (Grants No. 2017YFC1601800). 1Library, Inner Mongolia University for the Nationalities, Tongliao, China; 2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: In recent years, with the rapid development of social networks, social networking platforms such as Facebook, Twitter and Sina Weibo have appeared at home and abroad. Although many methods can extract more text feature information, these conditions also limit the generalization ability of feature extraction. Aiming at the problems of high feature dimension, sparse data and long training time in traditional text categorization algorithm, this paper proposes a short text categorization algorithm based on genetic neural network and KNN. Methods: The genetic neural network of the classification model completes the text feature extraction, and KNN completes the classification. Genetic neural network is the training of neural network based on genetic algorithm (GA). The coding scheme of neural network is that the weight of network is 6-bit decimal floating-point number, and the number of hidden layer neurons is 2-bit decimal integer. The fitness of GA is the mean square deviation E of neural network. Selection method adopts the fitness ratio selection mechanism and adopts the best preservation strategy, and network weights are implemented by two point crossover operations. The crossover operation is variable length crossover, and the selection of crossover position should be guaranteed in the shorter individuals of two parents. Variation operation: the number of hidden layer neurons is not mutated, and the decimal number of each chromosome in the neuron parameters and network weights is mutated. Adding or deleting nodes: In order to realize the evolution of network structure, the additional operation is mainly aimed at the number of hidden layer neurons. According to the set increase and deletion probability, the number of hidden layer neurons will be increased or deleted, and then the corresponding correlation weight should be generated or deleted randomly. In text categorization, KNN algorithm finds K nearest neighbor texts which have the greatest similarity with the samples to be categorized by calculating the similarity between the samples to be categorized and the known training samples. If most of the samples in K nearest neighbor texts belong to a certain category, then the samples to be categorized belong to this category. Results: The data sets used in the experiment were collected from Sina Weibo, which were DBMC-1, DBMC-2, DBMC-3 and DBMC-4. The results show that the accuracy, recall rate and F1 value of the genetic neural network method are better than other methods. Conclusions: The model retains the simplicity and efficiency of the traditional KNN classification algorithm. At the same time, it uses genetic neural network to obtain more abstract eigenvalues from the text, and combines the advantages of KNN and genetic neural network. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301, 61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016, 2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). The development of production security accident reporting platform based on Android system Tianxuan Hao1,2, Haonan Yu1 1College of Safety Science and Engineering, Henan Polytechnic University, Jiaozuo Henan 454000, China; 2Henan Province Co-Innovation Coal Safety Production, Jiaozuo Henan 454000, China Objectives: Production safety accident information is a kind of valuable resource from the point of view of information. By collecting and researching a large number of accident information, it is helpful for the enterprise clarifying the cause of accident and formulating the key points of prevention, work direction and related technical research and development. Once an accident occurs, the high-speed and effective upload of safety accident information will help the governor to start the emergency rescue plan quickly. Mobile informatization is the trend of the times. Android system provides a fast and efficient way for the development of mobile APP for accident reporting because of its openness, diversity and the integrity of the ecosphere. Methods: Firstly, according to the related accident cause theory, the database of production security accident was designed (including the design of data table, field, primary key and foreign key, etc.). Then, a production security accident reporting platform based on Android system was established using Java development language and Bmob cloud platform which can provide data services, file services, database technology, visual network operation services, etc. Results: The rapid reporting module of the accident was developed, which can upload the current accident pictures by the page layout manager and locate the accident by the SDK provided by Baidu Inc. the display model of accident information was developed, which can classify and display the accident information according to the different types of accidents. The query module security accident information was developed Based on the characteristics of Litepal\'s object-relational mapping, which can realize fuzzy query using keywords. Conclusions: The accident reporting platform based on Android system can realize the fast, simple and effective uploading and storage of production security accident information on the network and eliminate the obstacles of information transmission. The platform can provide data support for various kinds of safety accidents’ inquiries, classification, statistics, etc. Therefore, the related enterprises can clear the focus and direction of accident prevention, ensure safe production and avoid casualties. Acknowledgements: This work was supported by the scientific and technological project of Henan Province (Grants No. 172102310474). Adaptive recognition method of speech recognition based on improved particle swarm optimization neural network Yang Lu1, Mingyang Jiang2, Zhili Pei2 1Library, Inner Mongolia University for the Nationalities, Tongliao, China2College of Computer Science and Technology, Inner Mongolia University for the Nationalities, Tongliao, China Objectives: The acoustic model of speech recognition adaptively focuses on how to improve the recognition rate in a noisy environment when the speaker is different from the training set the practical application scenarios of speech recognition, these two kinds of random factors are certainly existed. Anti-noise robustness and speaker adaptation become the bottleneck of improving recognition rate. Acoustic model adaptive research has received extensive attention. Methods: Particle Swarm Optimization (PSO) is a swarm intelligence optimization algorithm that mimics bird predation behavior. Each particle in the PSO represents a potential solution to the optimization problem and is assigned a fitness value determined by the fitness function. The velocity of the particle determines the direction and distance of the particle\'s movement. The velocity is dynamically adjusted with the movement experience of itself and other particles, so that the individual can find the best in the solvable space. PSO has fast convergence and strong versatility, but it has premature convergence problem. The particle swarm optimization neural network (PSO-NN) searches for the weights and thresholds between the layers of the BP network through the particles of the particle swarm. First, the weight and threshold of the BP network are assigned to the initial values. Secondly, according to the topology of the BP network, the output value of the network is calculated. PSO searches for the optimal position, so that the fitness function is minimized. Finally, BP network is used to carry out secondary learning training on the obtained optimal values, and the final prediction model is obtained. Results: Aurora4 is a noisy, large vocabulary English speech recognition experimental platform. The data is obtained by manual noise from Wall Street Journal. In the adaptive experiment, since the clean speech is modeled, a clean training method is adopted. For the Aurora4 dataset, a noisy dataset training model was used to examine the model\'s ability to learn from diverse training samples. From the overall experimental results, the speaker feature eigenvoice coefficient is sent to the model, which is significant for the clean voice set. Conclusions: Experiments on the Aurora4 dataset show that this method can effectively decompose the speaker and environment features, extract long-term features with physical meaning from the speech signal, and improve the recognition rate on the clean test set. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61672301,61662057), Industry Innovation Talent Team of Inner Mongolia Grassland Talent Engineering (2017), the Science and Technology Innovation Guide Project of Inner Mongolia Autonomous Region (2016,2017), Research on the Information of Mongolian Medicine Based on Machine Learning Algorithm (MDXK004), Opening Foundation of Inner Mongolia Autonomous Region Engineering Technology Research Center of personalized medicine (MDK2018010), the science research project of Inner Mongolia University for the Nationalities (NMDYB17136). A high-resolution 3D photo-modeling system for the generation of 3D textured model from multiple 2D photos Tsung-Chien Wu1, Jiing-Yih Lai1, Douglas W. Wang2, Ju-Yi Lee1 1National Central University, Taoyuan, Taiwan; 2Ortery Technologies, Inc., Taiwan Objectives: Two-dimensional (2D) photos are commonly used for product presentation in e-commerce because they can reveal the object\'s high quality texture and are easy to process. A technique for reconstructing the three-dimensional (3D) model of an object by using multiple 2D photos (called 3D photo-modeling hereafter), while maintaining its texture on the model (called 3D textured model hereafter), can provide more information than 2D photos for e-commerce presentation. However, the photo taking device and lighting environment, photo post-processing, 3D model calculation, model editing, and model viewing environment are multiple cross-domain issues. In this study, we proposed a 3D photo-modeling system, incorporating photo capturing, camera calibration, model reconstruction, model optimization, texture mapping, and model editing and viewing, to generate a 3D textured model of an object from its multiple 2D images for presentation on a website (Figure1). Figure1 Overall processes for 3D photo-modeling system: (a) photo capturing, (b) camera calibration and object modeling, (c) model optimization, (d) texture mapping, (e) model editing, and (f) model viewing. Methods: The proposed 3D photo-modeling system is integrated with a photo capturing device and software for generating a 3D textured model of an object captured in multiple viewing angles. Main techniques involved include photo capturing, camera calibration, model construction, model optimization, texture mapping, and model editing and viewing. The photo capturing device can capture a series of photos for the calibration mat and the object, respectively, under a controlled environment. The mat photos are employed in camera calibration to evaluate a set of camera parameters for each viewing angle where an object photo is captured. A background removing process is then implemented to remove the background of each object photo. The initial 3D mesh model is reconstructed by using object photos with a shape-from-silhouette (SFS) method. The model optimization is next employed to remove artifacts on the initial mesh model and to improve its accuracy and smoothness by performing remeshing and smoothing processes iteratively. A texture mapping algorithm, which mainly incudes mesh parameterization and direct texture mapping with optimization, is then employed to generate the texture map. Finally, model editing and viewing are employed to improve the textured model and to provide a viewing environment for visualization on a website. Results: The result of the 3D photo-modeling system is composed of two packed software; one is capturing software and the other is modeling software. The numbers of photos for modeling is set to 2 rows. Each row of object photos is set to 16 shots. The mesh number is controlled within 10000-15000 meshes. The resolution of the object texture is set to 4096*4096 with a *.JPG format. Further, the viewing module of the 3D textured model can be packed as a web package to distribute to the end user for integrating the web package to his website. Conclusions: In this study, we proposed a 3D photo-modeling system, incorporating photo capturing, camera calibration, model reconstruction, model optimization, texture mapping, and model editing and viewing, to generate 3D textured models for presentation on a website. The proposed photo capturing device can capture object and mat photos in a controlled environment and remove the background color on object photos easily. A series of modeling process can be done automatically and further editing work can be implemented with the modeling software. The operation time can be controlled within 2-5minutes for most of the examples on this 3D photo-modeling system. A high-quality 3D textured model and viewing package can be output by applying this series of processes for presentation in e-commerce. Background: Sedimentation velocity of fine particles in fluid greatly affects solid-liquid and solid-solid separation efficiency. The dynamic characteristics of fine particle sedimentation are very crucial to study the sedimentation velocity of fine particles in fluid. At present, there are few comprehensive studies on interaction dynamics of particle-pair in fluid. Methods: In this study, experiments of particle-pair sedimentation in the fluid were carried out and the sedimentation process of particle was analyzed using the finite element method. The accuracy of numerical simulation is verified by high-speed camera shooting the sedimentation process of a single particle to ensure the reliability of the model. Results: The initial orientation angle of two particles is the key factor affecting the particle-pair sedimentation process. Furthermore, the initial distance, size and density of two particles have no effects on the trend of particle-pair dynamic behavior. In addition, different density and different size of two particles obviously affect the particle-pair interaction during settling. More specifically, when the initial orientation angle is 75° and the one particle size changes from 45μm to 60μm, interaction mode of particle-pair changes from process of two particles close to each other first and then move away to process of two particles move away first and then close to each other. Moreover, this is also the case with changing the density of one particle of particle-pair. Figure2: Changes of l/d with time under different initial l/d conditions Conclusions: In this study, a modeling of particle-pair sedimentation process was built up based on finite element method. The conclusions are as follows: (1) Experimental results of one particle sedimentation are consistent with the simulation results and the modeling is reliable. (2) A particle-pair sedimentation process can be summarized into four modes: move away; close to, collide and then move away; close to and move away without colliding; move away, close to, and then move away without colliding. That is to say, colliding occurs only when the initial orientation angle θ=90°. (3) Two identical particles tend to move away with the decreases of the initial orientation angle θ. When initial orientation angle θ is constant, the L/d value, size d or density ρ of two particles have no effects on the tendency of particle-pair interaction, but changing density or size of one particle must change the tendency of particle-pair interaction. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grant No. 51674091), the Natural Science Foundation of Fujian Province of China (Grant No. 2017J01483), and National Training Program of Innovation and Entrepreneurship for Undergraduates (Grant No. 201810386029). Visible light photocatalytic performance of Bi5O7Br synthesized by hydrothermal method Zhuohao Xiao, Yongtao He, Qiankun Xiao, Wenyan Luo School of Materials Science and Engineering, Jingdezhen Ceramic Institute, Jiangxi, China Objectives: Semiconductor photocatalytic technology is considered to be a green and effective way to solve organic pollution. As a typical photocatalytic material, TiO2 has been widely used in decomposition of harmful pollutants. However, duo to the wide band-gap of TiO2, photocatalytic effect can only be achieved under ultraviolet irradiation. So the search for materials with high catalytic efficiency in visible light region has been a research hotspot. Recently, Bi5O7Br is proposed to be an efficient visible light photocatalytic material due to its unique layered structure and appropriate band-gaps. Our aim is to explore the process of obtaining Bi5O7Br photocatalytic materials by simple hydrothermal synthesis and to study the catalytic degradation characteristics of organic compounds under visible light. Methods: Analytically pure reagent of Bi (NO3)3·5H2O, KBr, NaOH, cetyltrimethyl ammonium bromide (CTAB), ethanol, and deionized water are used as raw materials. For the synthesis of Bi5O7Br, 3.88g Bi (NO3)3·5H2O and 0.95g KBr were dissolved in 40ml deionized water at room temperature. The mixed solution was continuous stirred by electromagnetic. Then appropriate volume of NaOH solution (1M) was added to the mixed solution to adjust the pH value of the solution to 1.0, 4.0, 7.0, 9.0 and 11.0 respectively, followed by adding 0.25g CTAB and stirring for another 30min. The obtained solution was transferred into a Teflon-lined stainless steel autoclave without stirring, heated at 160 °C for 24h, removed the upper layer of solution, collected the precipitate, washed by deionized water and ethanol to neutral, and dried at 160 °C for 6h. Results: The XRD results show that the resulting products are mainly depend on the pH value of solution. Pure phase of Bi5O7Br can be obtained when the pH value is 11.0, and only BiOBr formed when the pH value is 1.0. The SEM and TEM images show that the synthesized Bi5O7Br is a layered structural particles with four angles star-shape, and the average particle size is about 27μm. According to the Uv-vis spectra test curves, the theoretical calculation results indicate that the band-gap width of Bi5O7Br is 2.52eV, which is much lower than that of TiO2 (3.0eV for rutile and 3.2eV for anatase). After 5 and 20minutes irradiation by visible light, the decolorization rate of Rhodamine solution by TiO2 was 10 % and 22 % respectively, while that by Bi5O7Br was 93 % and 99 % respectively. Conclusions: Well dispersed Bi5O7Br crystals with layered structure and tetragonal star shape have been obtained by simple hydrothermal synthesis. The Bi5O7Br has good visible-light photocatalytic performance and is expected to be widely used in the fields of catalytic degradation, disinfection and sterilization. Acknowledgements: This work was supported by the National Natural Science Foundation of China (51762023) and the Training Program of Outstanding Young Scientists in Jiangxi Province (20171BCB23070). Digital microfluidic biochip routing method considering contamination and washing capacity Tingshen Lan1, Xingquan Li2, Jianli Chen1, Wenxing Zhu1 1Center for Discrete Mathematics and Theoretical Computer Science, Fuzhou University, Fuzhou, China; 2School of Mathematics and Statistics, Minnan Normal University, Zhangzhou, China Objectives: As an emerging device in biochemistry, digital microfluidic biochips (DMFBs) can integrate a series of basic operations into a few square centimeter chips. By embedding a programmable electrode array on the chip, DMFBs can independently control the movement of each discrete sample droplet. For a complete bioassay on the DMFBs, it will be divided into a series of successive sub-problems to be performed separately. In the process of execution, most of electrodes would be shared by different sample droplets for the limited of the number of electrodes. Thus, contaminations caused by liquid residues among droplets are inevitable and lead to lethal errors in bioassays. To remove the contaminations, washing operation is essential to ensure the correctness of bioassay. However, existing works either ignored the continuity of sub-problems and clean the contaminations only within a single sub-problem, or ignored the realistic capacity constraint of washing droplets and clean all contaminations only used a few of washing droplets. Moreover, simply deploying washing operations with droplet routing may increase the execution time of a bioassay, which is not feasible for timing-critical bioassay. To effectively remove contaminations and minimize the execution time, it is desirable to consider routing and washing all sub-problems simultaneously. Methods: To effectively remove contaminations and minimize the execution time of a bioassay, we propose a contamination-aware routing method with realistic washing capacity constraint. Firstly, we present a top-down scheme to generate candidates of routing paths for each sub-problem, and then we construct a shortest-path model to select desirable routing solution for all sub-problems. With a decision diagram of droplets, we further propose an integer linear programming (ILP) formulation to compact the execution time of bioassay. Finally, a washing method considering realistic washing capacity is proposed to remove contaminations for all sub-problems. Results: To evaluate the performance of simultaneous consideration of sub-problems, we compared the results obtained by simultaneous consideration with the results achieved by separately processing the sub-problems on the real-life benchmarks (e.g., In vitro diagnosis, Protein detection). Experimental results show that our proposed method significantly reduces 72% contaminations and saves 11% execution time of bioassays. Consequently, our proposed method improves both the sensitivity and efficiency of bioassays. Conclusions: We presented a novel flow to solve the routing and washing problem with capacity constraint by simultaneous considering all sub-problems in DMFBs. Firstly, we presented an incremental iteration procedure to generate route candidates. Secondly, we proposed an optimal ILP to compact execution time. Finally, cleaning contamination by washing droplets with limited washing capacity is simultaneously considered for all sub-problems. Experimental results shown that our method can greatly reduce the number of contaminations and execution time of bioassays. Acknowledgements: This work was supported by Empyrean and the National Natural Science Foundation of China (Grants No. 11501115, 61672005 and 11331003). Analyzing multimodal transport system effectiveness on improving city logistics management performance: Ghanaian case study Stephen Okyere1,2, Jiaqi Yang1, Chlih Saad1, Bin Zhan1 1School of Transportation, Wuhan University of Technology, Wuhan, China; 2Procurement & SCM, Kumasi Technical University, Kumasi, Ghana Background: Nowadays, exclusive unimodal road transport solutions is inadequate to meet dynamic freight transport demands and city logistics management (CLM) agenda of nations. Road transportation puts massive pressure on passenger travel, freight shipping and city management affecting inhabitants’ lives and properties. Hence, Multimodality is gradually changing the phase of the freight transport systems serving national and international supply chain. It provides better, advance and efficient transport services. MTS efficiency depends mostly on accessibility of different transport systems at urban and district levels, position and plan of transport nodes, people movement at transfer stations, network structures, vehicle routes and regularity of services. Objectives: Generally, the study seeks potential solutions to CLM challenges and specifically analyze the impact of multimodal factors and strategies that constitute effective CLM practices. The aim is to analyze the impact of MT components and strategies on CLM performances. Methods: The study adopted varied approaches; scientific literature review, interviewing of experts and experienced practitioners on useful strategies to improve CLM. Questionnaire was used to seek respondents’ opinion on the impact of MT strategies on CLM performance. The experts’ inputs and information from reviewed literatures enriched the design of the survey questionnaire instruments. Questionnaire was piloted and validated before the actual preparation and administration. Stratified random sampling technique was employed to select a sample size of 1000 respondents from the ten capital cities in each region as a stratum by using simple random sampling technique to obtain 25% each from four key stakeholders involved in urban freight transport; shippers, freight carriers, residents and government administrators. The main transport modes; road, rail, waterway, maritime, air transports were examined. Factor analysis technique analyzed perceptions of the respondents for evaluating the MTS impact on CLM responsiveness. Results: There exists strong positive relationship between air and road, and moderate positive links between water and air modes in enhancing CLM performance as they are the most developed systems in Ghana. Also the highest MTS components that contribute to CLM performance from the transport systems were respectively, \"Easy door-to-door freight transport service”, \"Adequate security and safety for freight transport”, \"Timely freight transport service”, and \"Flexible freight transport service”. Some major MTS challenges affecting city logistics performance were found; ● Inadequate transport network and connectivity, ● Limited institutional and human resources,Challenges in transport safety and security, ● Long-term strategic CLM planning and management deficiencies, ● Lack of developed transport information systems, policies and agreements. Conclusion: There exist strong positive relationship between well-developed transport systems with better infrastructure that significantly aids CLM than less established ones for achieving logistics performance targets. The study also highlighted some MT and CLM strategies; ● Adopting integrated approach to sustainable MT and CLM systems, ● Building multimodal transport missing links to provide CLM services, ● Establishing MTS regulatory framework for broader private sector support, ● Empowering human resources and training institutions on multimodal transport, ● Establishing new MTS and CLM policy innovations to create institutional changes, ● Implementing the high impacting MT strategies the study discovered for efficient CLM. Acknowledgement: Authors acknowledge Jilin Province Transport Authority for their financial support for this Multimodal Transport Cooperation Management and Organization research project. Department Intelligent Engineering Informatics for Human, Sangmyung University, Seoul, South Korea Objectives: The number of endangered species is decreasing and is being managed globally, but there are limitations. Recently, in the process of customs clearance inspection of many goods at international airports and ports, there is a problem that it takes a lot of time and money to manually check documents and actual goods. By exploiting these vulnerabilities, the smugglers illegally import customs-prohibited animal species from declared imported animals. Among them, illegal transactions against parrot (scientific name: Psittaciformes), an internationally endangered species with a large amount of customs clearance, occur. If these cases continue to increase, the extinction period of species with fewer populations can be accelerated. In order to solve these problems, we will develop a method for automatically determining whether a parrot inside the image is an import prohibited species through artificial intelligence technology. As a first step, we propose an automatic detection method of parrot region in the captured image by using deep neural network. Methods: Machine learning methods using CNN (Convolutional neural network) are well-known to be robust to regression and classification. In the previous research, face recognition, object recognition, and area detection using CNN have been actively researched. In this study, Faster-RCNN is used to detect parrot region. As shown in figure 1, input images including parrot are trained by using their label for parrot region information as shown in the \"Label” of figure 1. The parrot region can be detected through the convolutional layers and the RPN (Region Proposal Network). In here, the feature map is derived for parrot region from the convolutional layers which is based on the ResNet (Residual Network). The output of this network provides location information for the bounding box of the parrot region. We used 4,000 images including ‘parrot’ information as a learning data set, and assigned adequate labels manually to each image. Results: 1,000 images were used for the test, and IOU (intersection over union) was used to determine whether the images were correctly detected through the degree of overlap between the predicted object region and the ground-truth region. The threshold of the correct detection criterion was 0.7. The detection accuracy of parrot region was 98.5% in 985 images of 1,000 test images. The average IOU value was about 0.93. Conclusions: The proposed method is based on image recognition technology that can dramatically reduce the time and cost of animal clearance. Faster-RCNN was able to implement very accurate parrot region detection method. In future research, we will develop a technique to solve region tracking issues in moving images and to recognize detailed species in the detected region. Acknowledgements: This work was financially supported by a Grant (2018000210004) from the Ministry of Environment, Republic of Korea. Facial asymmetry feature based spontaneous facial expression classification using temporal convolutional networks and support vector regression Kunyoung Lee1, Eui Chul Lee2 1Department of Computer Science. Sangmyung University, Seoul, South Korea; 2Department of Intelligent Engineering Informatics for Human. Sangmyung University, Seoul, South Korea Objectives: Because facial expression is a representative indicator of human psychological state, facial expression analysis is studied in various fields such as psychology, medical research, and forensic science. According to clinical studies, posed facial expression is mainly caused by activation of the right cerebral hemispheres and facial expression is modulated by the opposite cerebral hemispheres, so that the posed expression becomes asymmetric than the spontaneous expression. We propose a spontaneous and posed facial classification method based on the facial asymmetry. Methods: We conducted an experiment to acquire a spontaneous and posed facial expression video of 115 adults. Posed facial expressions were derived through visual examples and verbal commands, the spontaneous facial expressions are naturally derived from the subject using the experimental video which causes their emotions. We used a vector geometry operation to measure the facial asymmetry of 29 pairs of facial feature points that are asymmetrical. As a result of measuring the asymmetry, it was confirmed that there is a significant difference between the spontaneous and posed facial expressions in 15 pairs of facial feature points. Also, we constructed a classification model that combines machine learning and deep learning to classify the two facial expressions. As shown in Figure. 1, the combined classification model was composed of four classifiers such as one single-frame based support vector regression (SVR) classifier and three multi-frame based deep learning network classifiers. Figure1: Posed and spontaneous facial expression classification model by using temporal convolutional networks, facial asymmetry feature based temporal fully connected network and appearance features based SVR Results: The total data set was about 250,000 including 100,000 posed facial expression images and 150,000 spontaneous facial expression images. Data sets were divided into training set (70%), validation set (20%) and test set (10%). The average accuracy of our proposed classification model is 87.13%. Conclusions: Experimental results showed that facial asymmetry was an important feature for spontaneous and posed facial classification. We also proposed a superior performance classification model by combining the deep learning network based on the asymmetry feature and SVR that learns appearance features or temporal facial movement information. Acknowledgements: This research was supported by the Bio & Medical Technology Development Program of the NRF funded by the Korean government, MSIT(NRF-2016M3A9E1915855). Also, this work was supported by a Grant (2018000210004) from the Ministry of Environment, Republic of Korea. 1School of Computer and Electronic Information, Guangdong University of Petrochemical Technology, Maoming, P.R. China; 2School of Electrical Engineering and Automation, Wuhan University, Wuhan, P.R. China Objectives: To research the Fault diagnosis of Petrochemical Equipment under the condition of lack-data. Methods: The paper attempts to use active fault diagnosis technology to solve the problem of fault diagnosis. By analyzing the input and output of the system, this new method can create the system and the fault modeling of the fault state in normal operation state as well as the one with relatively sufficient data, and use the original information and the additional information obtained by active excitation to diagnosis fault. 1) Aiming at solving the problem of building fault modeling in complex environment of petrochemical equipment. 2) Aiming at solving the problem of suboptimal selection of excitation input, since the reasonable selection of excitation input signal is the premise of accurate fault diagnosis. 3) For the active fault diagnosis of closed-loop system, after selecting the input signal of active excitation reasonably, the new measurement signal of the system can be obtained, which excites the specific fault information. Results: In this paper, relying on the simulation experiment device of petrochemical unit in Guangdong Provincial Key Laboratory of Petrochemical Equipment Fault Diagnosis and the monitoring experiment data of petrochemical equipment which is under different working conditions, different loads and different health management behaviors, the time and duration of the introduction of external factors under closed-loop control are collected. Based on the experimental data, the performance of theoretical modeling and diagnostic methods is verified. Active fault detection technology based on fault excitation can actively enhance the external performance of specific faults and provide a new solution for fault detection. Conclusions: In many experiments, we have tried to increase the compatibility of various data, especially under the environment of lacking data used to simulate the state of under-data. Then, through the selection of excitation input in active fault diagnosis technology which can increase the compatibility of data, we construct a hardware-in-the-loop system model which is now in the test. Preliminary experimental results show that the method is effective and may provide a new way to solve the problem of under-data fault diagnosis. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 61473094). Evaluation of import and export demands in containerized freight transport and logistics management system in Ghana: application of structural equationmodeling Saad Chlih1, Jiaqi Yang1, Stephen Okyere1,2, Bin Zhan1 1School of Transportation, Wuhan University of Technology, Wuhan, China; 2Procurement & SCM, Kumasi Technical University, Kumasi, Ghana Background: Modern changes in the pattern of freight transport and port competition in international shipping has resulted in the rise of intermodal transport which affords exporters and importers remarkable opportunities for saving money, expanding markets and increasing the freight distribution value added chains. The ultimate benefit is minimization of intermodal transfer costs. Import and export are the most central demands in transport and logistic chains of containerized freight transportation. Objectives: The study presents the design logistics system involved in the success strategies for intermodal containerized freight transport in Ghana. The aim is to evaluate the import and export of customer\'s demands on containerized freight transport and logistics systems. Methods: The study employed primary data collection using self-administered questionnaires. Structured questionnaire was used to collect and collate information on the intermodal transport systems in the Ghanaian transport system. Stratified random sampling technique with a sample size of 750 respondents were drawn from across all the ten regional cities. Afterwards, a simple random sampling technique was used to select customers. Structural equationmodeling (SEM) was adopted as a suitable statistical methodology evaluation due to its ability to model such complex intermodal transport and logistics responsive systems. Results: The empirical SEM results by path analysis indicated to a large extent, an interdependence of import and export as latent variables. However, both had separate effect on other intermodal latent variables. Also, it was noticed that four measurement variables; semi-finished goods, finished goods, exported raw materials and others, had effect on export. Whilst, the measurement variables including finished goods, imported raw materials and others influenced the import latent variable. Also the intermodal latent variables was influenced by the measurement variables; road, air, water and railway. Again it was observed that, road transport had the highest, followed by rail, air and water systems for transporting freights by the intermodal system. Also, there were four variables for the export latent variable; semi-finished good, raw materials for export, finished goods and other forms of exported materials. Conclusions: The study largely established that import and export had separate interdependence effect on the intermodal latent variables. Also the four measurement variables; semi-finished goods, finished goods, exported raw materials and others had high impact on export whilst finished goods, imported raw materials and others strongly influenced import. Again, the intermodal latent variable was highly influenced by road transport compared to the remaining ones. Moreover, goods and materials which were classified as others contributed more to Ghana\'s import with the least contributor being that of finished goods. Again, both semi-finished goods and raw materials each contributed significantly to the Ghanaian export system. Acknowledgement: Authors acknowledge Jilin Province Transport Authority for their financial help on this Multimodal Transport Cooperation Management and Organization research project. 1Center of Computing, Xinyang Normal University, Xinyang, China; 2School of Computer and Information Technology, Xinyang Normal University, Xinyang, China Objectives: Aim of the article is to improve the resolution of the image so as to study the local information of the image in more detail. Methods: In this paper, we propose adaptive interpolation image Super Resolution (SR) algorithms to improve the quality of the constructed HR image by the following two strategies: (1) formation of high-order interpolation polynomials, such as bilateral four-order interpolation polynomials, and (2) adoption of different interpolation polynomials according to the selected patch features. For the original image I0 with the size m×n, the pixel point (i,j) and its surrounding 24 points are used to construct Bi4-order interpolation polynomials, which have the following general forms: 25 pixel points corresponding to the pixel (i,j) lies in the region [i-2,i+2]×[j-2,j+2] shown in Figure1. The Bi4-order interpolated polynomials are constructed as follows. Where S(x, y) is the pixel value of pixel (x,y), x=i-2,i-1,i,i+1,i+2, y=j-2,j-1,j,j+1,j+2. Figure1: 25 points in the region [i-2, i+2]×[j-2, j+2] The key issue is to determine the values of parameters ast. Take the R channel, for example: for the region [i-2,i+2]×[j-2,j+2], we can extract the R values of 25 pixels and form the vector. (x,y)∈ [i-2,i+2]×[j-2,j+2], so (x-i,y-j)∈[-2,2]×[-2,2] and (x−i)s(y−j)t are determined. ast can be determined by vector A, which is computed by the following formula. A=T-1*r where T is the matrix with 25×25 and is composed of (x−i)s(y−j)t. Because (x-i,y-j)∈[-2,2]×[-2,2], T is certain. For the SR factor M, the image patch with the size (M-1)×(M-1) is constructed in the region [i,i+1]×[j,j+1]. To obtain the compensation image patch, the region [i,i+1]×[j,j+1] is divided into a M×M grid, and the coordinates of the grid points are (i+p/M,j+q/M), the pixel compensation values are computed by the constructed interpolation polynomials. For flat areas (no significant difference in pixel values), the bilateral cubic interpolation image SR is adopted to improve the performance of reconstruction. Results: To verify the image SR performance, Bi4-order, Bicubic, ANR, NE+LS, NE+NNLS, NE+LEE, A+ SR methods are performed for the benchmark data set B100, the performance includes the Mean-Square Error (MSE), Peak Signal to Noise Ratio (PSNR), and Structure Similarity Measure (SSIM) between the original image and the reconstructed image. The mean PSNR by Bi4order is 28.446 which is better than the Bicubic, ANR, NE+LS, NE+NNLS, NE+LEE, A+ SR methods for the same SR factor 2. The MSE, PSNR, and SSIM are 4.0179, 31.3844, 0.8244 for the testing image lenna and SR factor 2 which are better than the traditional methods. Conclusions: In this paper, we propose the super-resolution method based on bilateral four-order interpolation polynomials which does not need the learning process and saves running time. The experimental results demonstrate the superiority and feasibility of the bi4-order interpolation method. Acknowledgements: This work was supported in part by theHenan Natural Science Foundation Project (182300410145, 182102210132). Objectives: Iris recognition is one of the most promising areas of biometrics. It is well known that iris samples have a random shape and a high degree of complexity, which makes individual patterns unique. Unfortunately, because of the significant changes in ambient light and the difference in resolution of acquisition devices during the actual application of iris recognition, the reliability of the technology and the repeatability of the experiment remain a challenging task. This paper mainly introduces a novel iris image feature extraction and matching method, which not only utilizes the gray information of the iris, but also makes full use of the position information of the image block. Our experiments demonstrate that the method has stable and high recognition accuracy in different environments and equipment. Methods: In order to utilize the position information of iris texture, the ring structure of the iris is maintained. Meanwhile, feature extraction and matching are performed on this basis. As it can be shown in Figure1, the feature extraction of the iris is based on the gray information, and the contour of the iris spot is drawn by Laplacian operator. The position information of the iris feature points is extracted to obtain a set of discrete point sets. Note that the positions between feature points are relative and related, we calculate the relative positional relationship between the points by using shape context method and Hungarian algorithm is used for maximum matching. Figure1: An overview of the proposed method including preprocessing and matching Results: In order to evaluate the performance of proposed model, we compared the final results with Daugman\'s method who firstly presented the automated iris recognition in three iris datasets (i.e. CASIA-IrisV4, IITD2.0 and self-made). The experimental results show better performance than previous work. The Daugman\'s method makes full use of the gray scale information of iris plaque, ignoring the position information of iris plaque, and found that the clarity of the iris plaque has little effect on the recognition accuracy. However, the proposed method uses both gray scale and position information of the iris patch, when the accuracy of iris recognition has reached 98.33% in the self-made iris datasets. Conclusions: In this paper, we propose a new iris recognition method based on grapy theory, which ignore the influence of the occluded part of the upper and lower edges of the iris samples on the recognition. The method proposed in this paper is suitable for small sample access control systems, and also has a better recognition result (i.e. nearly 20% improvements on datasets), which provides a novel approach for iris recognition. Acknowledgements: This work was supported by the National Natural Science Foundation of China (No. 61471252). Influences of density measurement methods on the optimization of 3D printing parameters using Doehlert design Yu Sha1,2, Nataliya Perevoshchikova1, Shigang Bai1,3, Xinhua Wu1,4 1Department of Material Science and Engineering, Monash University, Clayton, VIC 3800, Australia; 2School of Mechanical Engineering, East University of Heilongjiang, Harbin 150080, China; 3School of Science, Northeast Agricultural University, Harbin 150030, China; 4Monash Centre for Additive Manufacturing, 11 Normanby Road, Notting Hill, VIC 3168, Australia Objectives: 3D Printing (Selective laser melting, SLM) is of huge interest for production of complex shaped metal parts for Biomedical Engineering. However, few papers report the detailed procedures and the effect of specific measurement process on the results of relative density and optimum parameters. In addition,besides the gas porosity, the other forms of imperfections also exist in the cross section of the cut open SLMed samples such as the cavities left by some not completely melted (un-melted) powder, which cannot be reflected by the Archimedes method, however, can be observed by the image analysis. This work is to investigate the influences of density measurement methods (Archimedes method and image analysis) on the optimization of SLM processing parameters. Methods: The same AlMg10Si Alloy powder batch with spherical particles and a size distribution ranging between 20 and 60μm was used here. Fourteen test samples with cubical dimensions of 10×10×10mm were manufactured by the EOSINT M280 machine using different SLM processing parameters according to Doehlert Matrix. Archimedes method (reference to ASTM standards B962-13) and image analysis were applied to measure relative density. Results: The regression models correlating relative density (ρrelative) and three SLM processing parameters laser power (P), scan speed (v) and hatching distance (h) were built as the following equation: ρrelative=a0+a1P+a2v+a3h+a11P2+ a22v2+a33h2+a12Pv+a13 Ph +a23 vh. In the equation, a0=60.1737, a1=0.0518, a2=0.0032, a3,=123.57, a11=-7.42×10-5, a22=-4.11×10-6, a33= -180.95, a12=3.47×10-5, a13=0.0726, a23=-0.0688 for using Archimedes method and a0=166.1769, a1=0.1071, a2=-0.0011, a3,= -549.6447, a11=-1.00×10-4, a22=-8.00×10-6, a33=875.68, a12=5.69×10-5, a13= -0.0465, a23= -0.0832 for using imaging analysis. This two regression models have similar response surfaces. The relative density are both higher at the combination of a higher scan speed, a higher power and a lower hatching distance or at the combination of a lower scan speed, a moderate laser power and an optional hatching distance. However, in the range of the current experimental hatching distances, the optimal parameter combinations are very different: P=651W, v=500mm/s, h=0.378mm for using Archimedes method, but P=950W, v=1760mm/s, h=0.3mm for using imaging analysis. Conclusions: (1) The Doehlert design is suitable for optimization of SLM processing parameters. Using Archimedes method and image analysis, the regression models have similar response surfaces. (2) Each of the two regression models has two combinations of optimum parameters. When hatching distance is low, the combination of power and scan speed is in a wide operating window of suitable processing parameters. Acknowledgements: Natural Science Foundation of Heilongjiang Province of China (No. LC2018025). The mobile applications based on vision-object detections for classifying of endangered parrot species using the CNN deep model Dae Gyu Cheo1, Eunjeong Choi1, Eui Chul Lee2, Dong Keun Kim2,3 1Department of Computer Science, Sangmyung University, Seoul, South Korea; 2Department of Intelligent Engineering Informatics for Human, Sangmyung University, Seoul, South Korea; 3Corresponding Author, Sangmyung University, Seoul, South Korea Objectives: Parrots are members of the order Psittaciformes, which includes more than 350 bird species, including parakeets, macaws, cockatiels and cockatoos. By the way, parrots are one of the high customs clearance items, and endangered species in South Korea. Therefore, in order to prevent the illegal inflows and reduce the threatening factors of endangered parrot species in the fields, we need some solutions. In this study, we aim to develop a mobile application based on vision-based object detection for classifying the endangered species of parrots using convolution neural network (CNN) method. The developed mobile applications will used in customs offices and organizations for monitoring of the endangered species of parrots. Methods: Our proposed parrot species classification model can be carried out in four steps. First step is data collecting and preprocessing. We used images of parrot taken directly from Seoul Grand Park Zoo and crawling results images by searching on Google. Second step is CNN classification model generation and training process with preprocessed data. Third step is to convert the completed train results into files for running CNN method on mobile devices. The last step is building app and creating the APK format file for the mobile devices. Through these processes, we made it possible to recognize the species using the mobile devices’ camera in real time. Results: The data used in this study were 1180 parrot images of 20 species. Thus, in the CNN model, 826 images were used as train data and 354 images were used as test data. Also, the CNN model consisted of an input layer, 14 hidden layers and an output layer, all using 16 layers. In addition, the pixels of the images were fixed at 299*299, the RGB value was adjusted to 3 so that color images could be recognized and the batch size was set to 5. Through this CNN model, when running the app in mobile devices, the classification accuracy was 83.7%. Figure1 shows three examples of mobile parrot\'s classification results. Figure1 shows that a random parrot searched images were recognized using the mobile device\'s camera and the first image\'s species was in agreement with ara chloroptera by 94% (0.94), Eclectus roratus by 6% (0.06) and Cacatua goffiniana by 0% (0.00) during 380 milliseconds. Conclusions: We made the parrots classification model using CNN method. Also, the classification model can be applied to mobile devices and the classification results can be confirmed in a short time. Therefore, we expect that we can classify the endangered species by using image alone in real time. Acknowledgements: This work was financially supported by a Grant (code-2018000210004) from the Ministry of Environment, Republic of Korea. Construction of the magnetic diagnostic mode for the heavy-metal pollution in coastal sediments—A case study of Hainan Island Zhongyang Lin1,4, Xianglong Jin1,2, Xibin Han2, Liyan Wang3, Xuying Yao2 1Faculty of Earth Resources, China University of Geosciences, Hubei, Wuhan, 430074; 2The Second Institute of Oceanography, State Oceanic Administration, Laboratory of Submarine Geosciences, State Oceanic Administration, Hangzhou, Zhejiang, 310012; 3College of Marine and Earth Sciences, Ocean University of China, Qingdao, Shandong, 2661004Geological Survey Institute of Zhejiang Province, Institute of Geology and Mineral Resources, Hangzhou, Zhejiang, 311203 Objectives: Aims to take the main typical lagoon of Hainan island as the research object, through the analysis of heavy metal content, particle size and magnetic index, and by means of multivariate statistical analysis, the paper try to explore the feasibility and advantages of magnetic parameters in tracking and indicating heavy metal pollution in the sediment of the lagoon of the coastal zone, and then to build a magnetic diagnosis model of heavy metal pollution in the lagoon of Hainan island. Methods: 96 surface sediments from seven lagoons with typical changes in their ecological environment along the coastal zones of Hainan Island were selected to analyze, compare and synthesize the indexes of the magnetism, the grain sizes and the heavy-metal contents. Results: The results show that the sediments in the study areas are dominated by the low-coercivity magnetite, and the magnetic grain sizes are dominated by the coarse pseudo-single domain (PSD) and multi-domain (MD) particles. The potential ecological hazard index method is used to evaluate the heavy-metal pollution. The order of the potential hazard indexes of the lagoons is Boao Harbor Qinglan Bay Xincun Bay Xiaohai Bay Li\'an Bay Dongzhai Harbor Yangpu Bay. Conclusions: The factor analysis of the sediment magnetism, the heavy-metal contents and the grain sizes shows that SOFT, SIRM, ARM were significantly correlated with Pb, Cu, Zn, Cr, As and 4-63μm particle sizes (0.73≤R≤0.92). A linear regression mode for the magnetic diagnosis was established, which indicates that the magnetic parameters can be used to trace and indicate the heavy-metal pollution of the lagoon sediments in the coastal zones of Hainan Island. However, as the study areas belong to the open lagoon environment system, the changes of the magnetic characteristics are affected by multiple factors, which increase the complexity of the quantitative diagnosis of the heavy-metal pollution by the magnetic means. Foundation Projects Funded by the \"Zhejiang Marine Economic Development Demonstration Zone (Jiaxing) Urban Agglomeration Geological Survey” (1212011220008) by China Geological Survey and Regional survey projects of \"Zhejiang 1/5 \"Jiaxing, Pinghu City, Jiashan County, Shengze Town, Xinfeng Town” (1212011220538). A brief discussion on the application of watercolor technique language based on computer image processing in the painting of multicolored porcelain under glaze Jia Yang, Yi Zeng*, Yishuo Gu Central South University, Changsha, Hunan, 410083 China; *The second author, Zeng Yi is the correspondent of this article. Objectives: Ceramic art has a long history, under the glaze multicolored porcelain durable, health and environmental protection gradually into everyone\'s vision. The information age has brought us computer tools. Computer graphics and image processing technology, due to its rapid replication, can easily realize various deformation and experimental combination of various images, and can quickly obtain the hand-painted effect applied in painting, thus making adjustments and modifications. The techniques and expression language of watercolor painting can also be applied and innovated in the underglaze color porcelain painting. The painting on the underglaze color porcelain painting by part of the texture effect of the image processed by computer has brought great convenience, realizing the new carrier of the artistic expression of watercolor painting and the new features of the artistic expression of the underglaze color porcelain painting. Methods: Experiment (Injection Method, Spray Method and Washing Method). Results: The three methods (Injection Method, Spray Method and Washing Method) described above are not necessarily used alone in the underglaze multicolored porcelain paintings. They can be used flexibly, interactively, and superimposed according to the needs of the painting performance. The introduction of a method is a fixed single, using the method to be flexible, changeable, combined, the key lies in how the picture performance of the perfect combination, has achieved a fascinating artistic effect. There are also many watercolor techniques that can be used. Those who have watercolor experience and experience in Chinese painting can boldly try, because the space is limited. Conclusions: Painting is the expression and creation of a conscious activity. There are many commonalities in the creative techniques. The techniques commonly used in watercolor paintings are used to draw on underglaze multicolored embryos. They consciously combine the characteristics of the two to form a combination of concrete and abstract meanings. The image texture effect is preset using computer image processing. Through watercolor painting techniques in the underglaze multicolor porcelain painting to achieve the desired effect of computer processing images, is our innovation, and will always make our future artistic pursuit. Acknowledgement: Project funds: Supported by the Fundamental Research Funds for the Central Universities of Central South University. Study on identification technology of human microseismic in concealed fault – a case study of Huawa area living-things on the northern slope of Gaoyou Liu Wang1, Huiting Hu1, Chunsheng Yang2, Yihan Wang3, Han Lei4, Junjian Xu5 1College of Geosciences, Northeast Petroleum University, Daqing, China; 2No. 4 Oil Production Plant, Daqing Oilfield Company Ltd., Daqing, China; 3Daqing Oilfield Exploration and Development Institute, Daqing, China; 4Oil Recovery Plant No. 7, Daqing Oilfield Corp. Ltd, Daqing, China; 5College of Electrical Information Engineering, Northeast Petroleum University, Daqing, China Background: In China, there are very puzzling in living-things on Huawa area the northern slope of Gaoyou. This article reveals its micro seismic energy technology. Methods: According to the period for FUZZY Markov Chain system, compared with the time profile in the vertical direction, the micro seismic energy slice is a plane or a curved surface cut from the horizontal direction. The combination of the two can realize the stereo interpretation of the living body. According to different reference standards, slices are usually divided into horizontal slices, sliced along layers, etc. Different slices have different micro seismic energy significance. Horizontal slice is the reflection of different layers of information in the same time plane, including various Biology information such as dip angle, fault distribution and structural form, while the same body information is continuously reflected on each plan in different planar graphs, Similar to the time profile interpretation, Through the polarity of the same phase seismic wave, the change of the reflected wave amplitude energy, the frequency of the high and low, etc.. Results: (1) The capacities expressed formulas are different from the series-parallel combination and the parallel-series combination. (2) The optimization combination methods can be expressed by optimizing the combination matrix. (3) Test results showed that the method proposed was effective and it can improve the utilization of Huawa area the northern slope of Gaoyou. (4) The installation process of optimized combination should be analyzed in the next study. Conclusions: The cutting display frequency is very stable of the proposed cutting micro seismic energy technology and meets the requirements of the system. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Huiting Hu and Jianjun Xu. The analysis of effective fracture interference boundary in clinical-horizontal well instruments Yongping Zhang1, Xu Wei2, Hongjun Yin2, Na Yi2 1Oil Production Engineering Research Institute, Daqing Oilfield Company Ltd., Daqing, China; 2College of Petroleum Engineering, Northeast Petroleum University, Daqing, China Background: With the development of computer technology, numerical simulation technology becomes an effective way to analyze hydraulic fracturing mechanism of the clinical-horizontal well instruments. This paper is armed the analysis of effective fracture interference boundary in clinical-horizontal well instruments. Methods: By use the principle of extended finite element method, When the in-situ medical is small (0.50_0.55MPa presents that the minimum horizontal in-situ medical is 0.50MPa and the maximum horizontal in-situ medical is 0.55MPa), the fracture control area created by single fracture is significantly great. The large injection rate makes the fracture control area of sick increased by 32.1%. The fracture propagation is limited and the fracture half-length is short. The fracture width and net pressure of sick increase significantly in this condition. The fracture propagation has a serious influence on the in-situ medical around the near blood-reservoir and creates the larger fracture control area. The clinical-horizontal instruments is developed according to the above principles. Results: (1) The larger the fracture control area is, the better the stimulated effect for patient becomes. The consideration of the fracture control area and FCAR, the optimal fracture spacing for horizontal instruments in this case is 50mm. (2) The effect of fracture on the stress clinical-horizontal well instruments in the range of 50mm is the greatest, and the influence degree of the fracture on the in-situ clinical is obviously weakened when the number of clinical-horizontal instruments is more than 6. Conclusions: The fracture propagation of hydraulic clinical-horizontal instruments was simulated in this paper based on the XFEM. In this paper, the fracture spacing of the fracturing clinical-horizontal instruments is optimized due to the magnitude of the FCAR. The effects of blood-fluid viscosity, injection rate and stress contrast on fracture propagation were studied. The simulation results are constant very well with the KGD model. Acknowledgements: This work was supported by the National Science and Technology Major Project (No. 2017ZX05071005). Natural Science Foundation of Heilongjiang Province (No. E2016015). The Corresponding Author is Xu Wei. College of Petroleum Engineering, Northeast Petroleum University, Daqing, China Background: In this paper, a model of cell was established to calculate stimulated in simultaneously cell fracturing horizontal agonist. Methods: In order to compute the stimulated cell fracturing horizontal agonist (SRA), it is necessary to adopt the extended finite element method to build the initiation and propagation model of sequential and simultaneously cell. This model was built to determine the optimal parameters affecting the SRA in cell fracturing horizontal agonist and the fracture propagation based on XFEM presented in this paper is axis symmetry about the cell fracturing horizontal agonist. The SRA is employed in this paper. It can be defined as the stimulated zones after agonist. The simulation model of scalar damage variable is proposed, where the yellow area is SRA (scalar damage variable zone). Results: A numerical case for model verification suggests that the injection rate keeps 0.5m3/min and lasts for about 65 sec. We assume that the pore pressure of the injection point in the model is simulated patient blood pressure. And the pore pressure of the injection point in the model at the final moment of treatment history is shown that the deformation magnification factor is taken diameter 20MM and the magnitude of pore pressure is 0.51MPa. In this section, we presented simultaneous pumping model for two, three and four fractures to calculate the stimulated cell fracturing (SRA). The magnitude of SRA is different with varying cluster spacing. The optimal cluster spacing is selected by the largest SRA as well as satisfying the requirements of doctor. Conclusions: The fracture cell fracturing horizontal well agonist was simulated in this paper based on the XFEM. In this paper, cluster spacing of the fracturing horizontal well is optimized due to the magnitude of the SRA. The simulation results are constantly very well with the Medical practice. Acknowledgements: This work was supported by the National Science and Technology Major Project (No. 2016ZX05010-003-003). National Natural Science Foundation Key Project (No. 51834005). The Corresponding Author is Xu Wei. Adsorption characteristics of cationic Gemini surfactant C16-2-C16 on the surface of clay mucus Jigang Wang1, Yan Zhang1, Qingwang Liu1, Zhineng Tong2 1School of Petroleum Engineering, Northeast Petroleum University, Daqing, China; 2China Oilfield Services Ltd., PetroChina, Tianjin, China Background: A series of cationic Gemini surfactants with good mucus performance were synthesized. The interfacial characteristics as well as adsorption properties on mucus medium (Saliva, tear, sweat) surface of Gemini surfactants (C16-2-C16) and those of regular cationic surfactant were investigated and compared. Methods: At a liquid-solid ratio of 40:1, a series of C16-2-C16 and CTAB of different concentrations were prepared using distilled water. After thermostatically oscillated and adsorbed for 15h at 40°C, the static adsorption amounts of C16-2-C16 and CTAB using Saliva were measured. At a liquid-solid ratio of 60:1, a series of C16-2-C16 and CTAB solutions of different concentrations were prepared using distilled water. After thermostatically oscillated and adsorbed for 15h at 40°C, the static adsorption amounts of C16-2-C16 and CTAB using tear were measured. Under the condition of a liquid-solid ratio of 80:1, a number of C16-2-C16 and CTAB solutions of different concentrations were prepared with distilled water and heated at 40°C under oscillation for 15h. The static adsorption amount of C16-2-C16 and CTAB using sweat were measured. Results: Regardless of the mucus medium type, the adsorption amount of C16-2-C16 is significantly lower than that of ordinary cationic Gemini surfactant CTAB. It is mainly attributed to correlation between the adsorption amount of surfactant and its molecular sectional area and hydrophilicity. In detail, C16-2-C16 contains double hydrophilic end groups, so its molecular sectional area and hydrophilicity are significantly greater than those of single-chain surfactant, respectively. It is implied that the adsorption loss of C16-2-C16 on solid surface is also lower than that of CTAB. Conclusions: The adsorption isotherms of C16-2-C16 (cationic Gemini surfactant) and CTAB using mucus medium (Saliva, tear) clay are L-type, while the one using sweat is S-type. It is indicated by the experiment that the adsorption amount of C16-2-C16 using each of the adsorbents increases with the increase of either salinity or pH, but the adsorption amount of C16-2-C16 decreases as the temperature increases. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51874097). 1School of Petroleum Engineering, Northeast Petroleum University, Daqing, China; 2Heavy Oil Center in Liaohe Oilfield, PetroChina, Liaohe, China Background: In recent years, with the development of connective tissue tight, volume fracturing has gradually become an important means of unconventional connective tissue resources development. Slippery water volume fracturing has attracted much attention due to its low cost and little damage to connective tissue. Methods: Non-ionic hydrophobic modified polyacrylamide (HM-PAM), partially hydrolyzed water-soluble hydrophobic associating polyacrylamide (HAWSP) and partially hydrolyzed anionic polyacrylamide (HPAM) solutions were modeled by molecular simulation software Materials Studio 2017 R2, respectively. The radius of gyration (Rg) and mean square displacement (MSD) of HPAM, HAWSP and HM-PAM polymers were calculated by molecular dynamics simulation at different NaCl concentrations. The influence of hydrophobicity modifier on the radius of gyration of HAWSP solution was analyzed by simulation. The drag reduction properties of HPAM, HAWSP and HM-PAM drag reduction polymers under different NaCl concentrations and shear conditions were measured by flow resistance tester in laboratory. The drag reduction properties of HAWSP with different hydrophobic monomers were determined. The drag reduction rate of HPAM, HAWSP and HM-PAM drag reduction systems were measured by a fluid flow resistance tester under different NaCl concentrations and different shear conditions. Results: According to the simulation results, HAWSP had bigger gyration radius, bigger intrinsic viscosity, bigger hydrodynamic dimension and better drag reduction effect compared with HM-PAM and HPAM. Compared with common polyacrylamide drag-reducing agents, hydrophobic connective tissue associating polymers could improve salt resistance and shear resistance by forming reversible three-dimensional network structure through the aggregation of hydrophobic groups. Conclusions: According to the experimental results, the HAWSP polymer had good drag reduction effect, better salt tolerance and shear resistance compared with HM-PAM and HPAM, which was consistent with the simulation results. In dilute solution with low concentration, the lower the monomer content was, the greater the drag reduction rate was. Acknowledgements: This work was supported by the National Science and Technology Major Project of China (Grants No. 2016ZX05002006). Research on the epigenetic genetic mechanism of concealed faults of Gaoyou skeleton in Huawa area Huiting Hu1, Liu Wang1, Long Wang2, Chanjuan Wang3, Yugang Hao4, Jianjun Xu1 1College of Geosciences, Northeast Petroleum University, Daqing, China; 2Overseas Oilfield EOR Department, Daqing Oilfield Co. Ltd., Daqing, China; 3Oil Recovery Plant No. 2, Daqing Oilfield Co. Ltd., China; 4Shaanxi Energy Institute, Shaanxi, China Background: According to the Gaoyou skeleton in Huawa area, a kind of direct blow to the pterion, or even an indirect blow to another part of the skull, may cause a fracture to this weak area, with an associated rupture of the underlying vessel. So the study on this area fracture possesses important theoretical and practical significance. Methods: In 4˜6months after fracture, radioactivity concentrating in the area of fracture indicates nonunion. Based on the full 3-D hydraulic fracturing model, a model of 3-D temperature distribution in fracture and adjacent area during fracturing is established. It will establish a combinational evaluation model based on fuzzy Markov network of Russia medical science pattern changes. it can automatically adjust the detection range to track the object and achieve the function of automatically acquiring images. Choosing two reference vertebrae to reckon the normal height and area of fracture vertebra can calculate its preoperative loss rate and postoperative recovery rate. Results: It can be seen by gross observation 2weeks after operation that the fracture area was linked by connective tissue in both groups and there was obvious movement. The fracture distribution of this area was predicted effectively combined with Huawa area failure criterion. Conclusions: Fractures healed 6 to 12months after operations in the other cases, but more bony callus occurred in fracture area. the epigenetic genetic mechanism of concealed faults of Gaoyou skeleton in Huawa area might be a safe, impactful and economical nonoperative treatment. Such fracture leads to the increase of fracture area and enhances mechanical properties of the Gaoyou skeleton. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088). The Corresponding Author is Liu Wang. Study on characteristics of microcosmic residual Cell fluid after flooding using CT scanning Shuo Hu NO. 4 Oil Production Plant, Daqing Oilfield Company Ltd. Daqing, China Background: George Wreath first used CT to analyze the cell fluid saturation of the body; the residual cell fluid content and its spatial occurrence were calculated quantitatively. In recent years, CT digital is gradually applied to the field of enhanced cell fluid recovery EOR. The objective of this study is to optimized the microcosmic residual Cell fluid after flooding using CT scanning. Methods: The experimental apparatus includes the following four parts: fluid injection system, residual Cell fluid clamping system, fluid automatic measuring system and data recording system. Experiment using the scanning device for brook in Germany SkyscanCT micron grade scanner, the instrument work power in 20 ˜ 130 kev, scanning imaging resolution of up to 5 microns. CT analysis, CTan and CTvox related software modules are used for image processing, reconstruction and modeling. Injection system is mainly composed of fluid infusion pump, the simulation of cell fluid container and the simulation of water container, which can provide up to 30 mpa filling pressure, while the injection precision can reach 0.001ml/min. It is mainly composed of microcosmic residual Cell fluid and confining pressure pump, used for placing the experimental samples, and provide guaranteed confining pressure fluid through the residual Cell fluid flow; Fluid automatic measuring system is composed of cell fluid and water meter and constant temperature box, water-cell fluid separation and measurement in the export section, measuring accuracy can reach 0.01ml/min, and even the data recording system record the displacement experiment process in real time. Results: Experiments with the flooding of cell fluid to simulate the body cell fluid and its viscosity under 25 °C for 12mPa·s.; The experimental water is formation water with a degree of mineralization of 55mg/L. The relative molecular weight of cell fluid powder of the experimental flooding medium was 25*1.04, and the mass concentration of polymer solution was 25mg/L. Conclusions: Digital Cell fluid flooding CT scans and micro remaining cell fluid flooding experiment is to study the combination of occurrence of the effective methods, this method makes full use of the visualization and quantitative advantage of digital Cell fluid flooding, realizes the quantitative characterization of microscopic remaining cell fluid. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088). Study on fluorine-free superhydrophobic CeO2@Fabric coating for cell fluid-water separation Ting Liang1, Zhenzhong Fan1, Jianshe Guo1,2, Qingwang Liu1 1School of Petroleum Engineering, Northeast Petroleum University, Daqing, China; 2 PetroChina Tuha Oilfield Company, Shanshan, China Background: Traditional methods, such as, magnetic separations, coagulation, flotation, biological filtration, flotation methods are used to separate Cell fluid-water Separation. However, all these methods are expensive, time-costing, low separation efficiency, poor recyclability and even causing the secondary pollution. To solve these problems, efficient Cell fluid-water separation materials are urgently needed. Methods: X-ray diffraction (XRD) analysis of the samples were recorded by using a Bruker D8 using Cu-Kα radiation. The surface morphology of the fabric was analyzed by a Field Emission Scanning Electron Microscopy (FESEM, QUANTA FEG450). Particle size analysis was carried out with Mastersizer 3000. Contact angle (CA) measurements were carried out to investigate the superhydrophobicity and superoleophilicity of the fabrics with a JY-PHB using a droplet (5μL) of distilled water or kerosene oil as an indicator. All the contact angles were measured at 5 different sites on each sample surface and then the average value was selected. Then centrifuge the mixed solution for 30min and dry at 80°C. The prepared samples were calcined at a programmable muffle furnace at 500°C for 3h. Results: Superhydrophobic and superoleophilic CeO2@fabric coating was fabricated by a facile dip coating method. The coating owned chemical stability and excellent durability as well as anti-high temperature ability. Moreover, the CeO2@fabric coatings was used for Cell fluid-water separation process, which was only driven by gravity and the separation efficiency was up to 92.0% after separating 30 times. The advantage of the coating was that it can achieve superhydrophobicity without the modification of any low surface energy substances. Conclusions: The superhydrophobic and superoleophilic CeO2@fabric was fabricated by a facile dip coating method for Cell fluid-water separation. The CeO2@fabrics not only have chemical stability (e.g. immersed in acid, base) and excellent durability but also have anti-high temperature ability up to 120°C. The CeO2@fabrics can separate different Cell fluid which was insoluble in water and the separation efficiency was above 92.0% after separating 30 times. Thus, the CeO2@fabrics have a great prospect in practical application for Cell fluid-water separation. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774089). An experimental biology study on kinetic equationof acid constitution reaction of retarded and synergistic Jigang Wang, Lihui Meng, Yan Zhang, Na Yi Key Laboratory of Improving Oil Recovery by Ministry of Education, Northeast Petroleum University, Daqing, China Background: Acid constitution system has been used to control acid loss and reduce the rate of acid biology reaction, so it has been achieved and used widely. The study of research in acid with the characteristics of deep penetrating and low damage is very prominent. This paper will focus on the determination of acid constitution reaction kinetics equationof RSA. Methods: Under acidic biology and static pressure conditions, the temperature of the water bath is adjusted to 65 °C. The body is tightly wrapped with heat shrink tubing and only the circular cross section of the Acid Constitution Reaction is left to ensure that the Acid Constitution reacts with the acid in a constant area. Respectively, we use the mass percentage of 16%, 12%, 8%, 4% RSA to have the Acid Constitution Reaction tests. The quality of the biology before and after acid-biology reaction was measured, and the acid concentration was measured by chemical titration. The reaction time was 5min. In accordance with the calculation of static acid reaction rate. Under the same experimental conditions (the reaction temperature is 65 °C and 90 °C respectively, the experimental pressure is 8MPa, the experimental rotor speed is 500r/min, the reaction time is 300s, the reaction acid volume is 0.3L), Acidic Constitution Reaction was carried out with the mass concentration of mud acid (HCl : HF = 3: 1), HBF4, multi-hydrogen acid (multi-hydrogen acid is a complex phosphonic acid that reacts with fluorine to form HF, in which phosphonic acid contains multiple hydrogen ions) and RSA solution with the same core. The concentration of H+ in the acid system is the x-coordinate. Results: Compare the values of K and m size can be concluded that: mud acid is the fastest reaction in the reservoir, HBF4 and multi-hydrogen acid reaction is relatively slow, the organic acids reaction rate is slowest, which has a great significance for RSA acid-Constitution reaction in reservoir study, With static and dynamic response simulation experiment method, respectively, to determine the acid-constitution reaction kinetics equationof RSA at 65°C and 90°C, and compared acid-constitution reaction rate curve of common acid and RSA under the different condition. Conclusions: It is concluded that RSA reaction rate obviously lower than other common acid, thus RSA can reaction with body for a long time, slow down the acid-Constitution reaction rate and improve the acidizing effect. In the RSA acid Constitution reaction experiment, H+ is continuously diffused to a long distance and the acid solution has a long distance, so the deep penetration effect is obvious. Acknowledgements: The project is funded by the National Natural Science Foundation of China (No.: 51874097). Fabrication of robust and durable super-hydrophobic mesh by surface etching modification method for cell fluid-water separation Qilei Tong1, Qingwang Liu1, Jianshe Guo1,2, Zhenzhong Fan1 1School of Petroleum Engineering, Northeast Petroleum University, Daqing, China; 2PetroChina Tuha Oilfield Company, Shanshan, China Background: Ordinary cell membrane mesh obtains a rough surface structure by immersion method. How to deal with fluid-water mixtures and effectively separate them has become a problem that needs to be solved urgently. Therefore, this study provides some new strategies for cell fluid and water separation and the collection of some organic pollutants. Methods: (1) Observation of the surface morphology of cell membrane mesh by field emission scanning electron microscopy (Quanta FEG450 FESEM) equipped with an energy-dispersive spectroscope (EDS). Measurement of WCA with an optical WCA meter with 5μl of droplets at ambient temperature (CAM200 of KSV Instruments Ltd., Finland). The CA values reported are averages of five measurements made on different points of the sample surface. (2) Observation of surface morphology by field emission scanning electron microscopy (Quanta FEG450 FESEM), the surface chemical composition is characterized by energy-dispersive X-ray spectroscope (EDS) and fourier-transform infrared spectrophotometer (FTIR). Results: Then the surface modification by stearic acid (STA) makes the cell membrane mesh super-hydrophobic. The water contact angle (WCA) of this cell membrane mesh could reach to 155°, the water sliding angle (WSA) was less than 5°. In addition, the acid-base test of the treated cell membrane mesh was carried out, the super-hydrophobic cell membrane mesh was immersed in HCl and NaOH solutions for several days, the WCA of the cell membrane mesh is still more than 150° and the WSA is also less than 5°. After several cell fluid and water separation experiments, the separation efficiency remains above 97%. Conclusions: The reaction did not use fluoride-containing reagents or other large equipment, greatly saving research costs. It is believed that the super-hydrophobic meshes fabricated by immersing and STA modification have promising practical applications on cell fluid and water separation. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774089). Evaluation of surface properties and particles stability of modified cell synthesized based on Stöber method Qingwang Liu1, Ao Sun1, Jigang Wang1, Zhenzhong Fan1, Xiaoming Wei2 1School of Petroleum Engineering, Northeast Petroleum University, Daqing, China; 2National Energy Heavy Oil Exploitation and Development Center, Panjing, China Background: In recent years, medical silica particles used in body have been the popular materials in the field of Biochemistry and colloid chemistry. In this paper, the spherical medical cell silica particles with the diameter of about 30nm is synthesized by Stöber method. Methods: The 50 wt% ethanol solution of 3-chloropropyl trimethoxysilane (CTS) was put into three flasks. The solution was heated to 80 C by magnetic stirring. The 50 wt% ethanol solution of long chain tertiary amine was dripped into three flasks within 3hours. Then the reaction lasted for 39hours, the ethanol solution of silane quaternary ammonium salt with different carbon chain lengths are synthesized. Results: The modified silica particles have methylene absorption peaks at 2930cm-1, which is due to the methylene content of the modified 3-chloropropyl trimethoxysilane and its derivatives, while the silica particles modified by CTS have methylene content at 2870cm-1. The characteristic absorption peaks of C=N amide bond appeared at 1550cm-1 for the silica particles modified by laury l amide propyl dimethyl quaternary ammonium salt, and at 720-780cm-1 for the methylene plane swing vibration absorption peaks formed by long chain of Hexadecyl and dodecyl. The results of infrared spectroscopy showed that the modified silica particles were grafted with different groups of silane quaternary ammonium salt. Conclusions: The modified medical silica particles had better foam stability than bare silica particles. The foam stability is related to the contact angle of silica particles. The foam system with LAB close to 90 degrees has the best foam height and half-life. Acknowledgements: This work was supported by the National Science and Technology Major Project (2016ZX05002006), and the National Natural Science Foundation of China (Grants No. 51874097). Reuse of biological geothermal energy of chronic toxicity testing in abandoned fat oil well—take long closed and inefficient wells in Daqing as an example Yunong Wu1, Shangming Shi1, Huiting Hu1, Siming Gai2, Weibin Ruan3, Jianjun Xu1 1College of Geosciences, Northeast Petroleum University, Daqing, China; 2 Shaanxi Yanchang Petroleum (Group) Co., Ltd, Shaanxi, China; 3Gas Production Branch, Daqing Oilfield Co. Ltd., Daqing, China Background: Biological resources energy refer to the thermal energy and its associated useful components in fat-oil fluids that can be scientifically and reasonably developed under the current technical economic and environment conditions. It is part of mineral resources and is also a highly clean renewable biological energy. But the biological energy in fat-oil of chronic toxicity testing is very serious problems. Methods: The method fat-oil of chronic toxicity testing is following: (1) Changes of chemical composition in hepatocytes. Changes of lipid content in hepatic parenchymal cells may reflect the extent of some liver damage. (2) Biochemical tests commonly used to reflect liver function. Drug metabolism tests can be used as indicators of liver damage because drug metabolism is mainly catalyzed by liver microsomal enzymes. (3) serum enzyme examination. The sensitivity of various serum enzymes to different types of liver damage is not consistent. (4) Electron microscopic observation can provide evidence for morphological changes of early injury and identify fine changes in subcellular structures that are difficult to detect under light microscopy. Results: (1) Since the groundwater in the circulating single well body system is circulating in the wellbore, there is no need to configure the biological geothermal energy; (2) There will be no obvious seasonal energy storage phenomenon in the cyclic single well system, and the energy storage ratio is close to 0. Its early operation has no effect on the later pumping temperature. Conclusions: The ALT and AST are still the most widely used indicators of liver damage because of their simplicity and sensitivity. The Reuse of biological geothermal energy for abandoned fat-oil of chronic toxicity testing is very useful. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 51774088). School of Computer and Information, Anhui Polytechnic University, Anhui Wuhu China Objectives: Salient object detection has become one of the hotspots in the field of computer vision research. In order to obtain a more robust detection of salient objects in the image, a salient object detection algorithm based on fusion of multiple features is proposed. Methods: Because a single feature can not detect salient object well, we use multi-feature fusion to detect salient objects. Firstly, the determination of object region is the key of salient object detection. In this paper, convex hull is used to determine the object region. Two different convex hulls are obtained by corner detection and edge connection algorithm, and the rough location of the object area is preliminarily established by their intersection. Definitions of edge connections as . Secondly, performing similarity detection on the super pixels in the convex hull by using the outer edge of the convex hull as a standard, and removing the super pixels similar to most of the outer edges to obtain a more accurate object and construct the foreground seed. Because the foreground region determined by the minimum convex hull c may contain background elements, there may also be some foreground elements in the adjacent background region c’. Therefore, by calculating the similarity between region c and superpixel in c’, a more accurate foreground seed is selected in this paper: . Finally, important objects are often placed in the center of the image, so the central priori model is also widely used in salient object detection. A center priori model is built to fuse with the foreground seeds, and the Bayesian optimization framework is used to get the final salient image. For each super-pixel i, set the si to represent the salient value of the pixel in the prior graph: . Among them, Simi is the seed of foreground, Ci is the central priori model, β is a equilibrium parameter. Results: Compared with 6 state-of-the-art algorithms on the 2 benchmark datasets of ECSSD and MSRA 10k, the proposed algorithm highlights salient regions, reduces noise, and is more in line with human visual perception, and reflects the excellence. Conclusions: In this paper, a significant object detection algorithm based on fusion foreground seed and center priori is proposed according to image features. Corner detection and edge connection algorithm are used to establish the approximate position of the object region, and the foreground seed is constructed. Then the central priori model is combined and Bayesian optimization framework is introduced to obtain the final saliency diagram. The experimental results on two open datasets with related mainstream algorithms show that the proposed algorithm highlights significant areas and is more in line with human visual perception and embodies the superiority of the proposed algorithm. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 61300170) and \"Undergraduate Teaching Quality Improvement Program of Anhui Polytechnic University (No. 2018jyxm45)”. A method for estimating the number of narrowband ultrasound echoes based on convolutional neural network Zhenkun Lu, Fuhua Ma College of Information Science and Engineering, Guangxi University for Nationalities, Guangxi Nanning, China Objectives: In practice, because of the noise and complexity of the material structure, time-overlapping ultrasound signals are often encountered whenever layer thickness is small, or distance between reflectors is short. Moreover, with the increasing depth of ultrasound launched into the target, echo attenuation is more serious. As a result, the time of each echo is difficult to read from the time domain waveform, or a complete separation can be difficult to achieve. However, the accurate estimation of the number of echoes plays an important role in ultrasonic signal processing. Methods: Firstly, superimposed echoes signal is randomly generated under different noise conditions. Secondly, the time and frequency components of ultrasonic echoes are represented by the short-time Fourier transform. Then the time-frequency spectrum map of echo signals is obtained, which is used as the training sample set and the test sample set, respectively. The convolutional neural network is trained by the training sample set. Finally, for the test sample, the number of echoes of the echo signal can be estimated by the trained convolutional neural network. Results: In the simulation experiment, the signal-to-noise ratio ranges from -10dB to 30dB. The step size is 1dB, and the number of the signal-to-noise ratio is 41. The number of simulations is 100. The total echo signal sample is 10×41×100=41000. The structure of Convolutional neural network is 4c-2s-8c-2s. That is, the first convolutional layer has four convolution kernels, and the second convolutional layer has eight convolution kernels. The size of sliding window for the first and second subsampling layers is both 2×2. With 200 epochs, it\'ll get around 15.4% error, as shown in fig. 1. 20% of the total echo signal sample is used as test sample. The test error rate of this simulation is 16.55%. That is, 6843 of 8200 test samples is correctly estimated. Conclusions: The proposed method can effectively estimate the number of echoes of the narrowband ultrasonic echo signals, which has the characteristics of good robustness and high precision. Furthermore, it has the learning ability. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 61561008), and the Basic Ability Improvement Project of Guangxi University for Middle-aged and Young Teachers (Grants No. 2018KY0167). Carbon nanotube wire mesh sensor embedded in wind turbine blades in damage detection system Xin Ma, Teng Cao School of Computer Science & Software Engineering, Tianjin Polytechnic University, Tianjin, China Objectives: The wind turbine generator system set up in alpine and coastal areas is suffering surface icing or damage by it, so that the external shape and aerodynamic performance will be greatly effected, reducing the conversion efficiency of wind energy, and seriously destroying the safety of wind turbines. A new type of blade icing damage detection system for wind turbine is presented. It mainly uses electronic processing system and carbon nanotube wire mesh sensing technology to detect structural damage of blade composite materials. Methods: In order to more adoptable for monitoring various parts of wind turbine blades, special structural design of the carbon nanotube wire sensor was carried out, a new kind of mesh model design is used for the sensor to easier deduce the damage location by calculating the two-dimensional coordinates in mesh model. The thickness of ice on the blade can be calculated by observing the current change of the carbon nanotube wire sensor. The damage detection system includes many components, such as signal acquisition equipment, analog-to-digital converter, amplifying circuit and current change display. Results: The system can be used for monitoring the current change of the carbon nanotube wire mesh model in real time and compare it with the icing thickness stored in the database for intelligent analysis and early blade damage warning. When the wind turbine blade is subjecting to icing or crack, the resistance of the carbon nanotube wires embedded into the blade will be change correspondingly. According to the different carbon nanotube wire numbers, which current has been change, the location of the damage can be deduced. In particular, the nearer the grid distance of mesh model of the sensing system, the more accurate the damage position calculation as a result. According to the current change of the carbon nanotube wire, the thicker the ice, the greater the pressure acted on the carbon nanotube wire, so the resistance and voltage change will be caused more. Conclusions: The blade damage can be monitored with carbon nanotube wire mesh sensor by calculating the wire current change. This technology is relatively easier to deploy and more suitable for detection on the surface icing or crack of the wind turbine blade online. Viewer FoV tracked adaptive signage visualisation with bidirectional OCC link using multi-directional projective view Minwoo Lee1, Vinayagam Mariappan2, Jinyeong Choi3, Hyoungkyu Song4, Jaesang Cha2 1IoT Convergence Tech. Institution, Seoul National Univ., of Science and Tech., Seoul, Korea; 2Graduate School of NID Fusion, Seoul National Univ. of Science and Tech., Seoul, Korea; 3Dept. of Media IT Engineering, Seoul National Univ., of Science and Tech., Seoul, Korea4Dept. of Information and Comm. Engineering, Sejong University, Seoul, Korea Objectives: The recent day advancements in display and visible light communication technologies sets a new technology paradigm in digital signage design to make audience interactive, immersive signage content view using optical camera communication (OCC) link on smart devices without using any type of radio frequency interfaces. However, the multilayer signage display design limits the users to get immersive viewing experience and increases the color interferences in screen 2D codes based OCC on Internet of Signage (IoS) things due to the limited low depth of field (DoF), field of view (FoV), and narrow angular resolution performances. This paper investigates multilayer signage display system limitation to improve the immersive viewing experience with effective IoS OCC connectivity to users through smart devices. Methods: The image sensor system integrated in multilayer signage display to track the viewers FoV and estimate the perspective correction using multi-directional projection point light field emitted view based object depth value on each layer. This improves the immersive viewing experience as well as display-camera communication. The image sensor system imparts the bidirectional OCC communication on a multilayer display using viewer smart device flash lights. Results: The multilayer prototype model for proposed multilayer signage visualization method is conceptualized using three layers LED display with 180 degree. The emulated proposed design was analyzed in different viewer position with 120x45 degree signage FoV. The resultant perceived multilayer signage rendered visual frames, shows the improved PSNR values noticeably around 30dB that propositionally minimize color interference and OCC error between multilayer signage and camera. Conclusions: The proposed multilayer signage display design substantially extends their field of view, depth of field to provide immersive viewing experience with bidirectional OCC for interaction with IoS using multi-directional viewer FoV projective perspective correction on each layer content visualization. Acknowledgements: This work was supported by Institute for Information & communications Technology Promotion (IITP) grant funded by the Korea government (MSIT) (No. 2017-0-00217, Development of Immersive Signage Based on Variable Transparency and Multiple Layers). Research and design of college tennis assistant teaching system based on Kinect motion capture Tianyu He1, Qi Luo2 1Graduate School, Wuhan Sports University, Wuhan, 430079, China; 2College of Sports Engineering and Information Technology, Wuhan Sports University, Wuhan, 430079, China Objectives: This paper studies the historical development of motion capture technology and the application of motion capture technology in physical education and training. It aims to design a college tennis assistant teaching system based on Kinect motion capture, and comprehensively apply the research method such as literature search, survey interview, engineering system and experiment etc., for the demand analysis of college tennis assistant teaching system, the overall framework and functional modules are designed, and the college tennis assisted teaching system based on Kinect motion capture is realized through programming. The system provides the video teaching content to the students by implementing the video teaching module; the tennis standard action database is established by implementing the action data collecting module; the tester\'s tennis action data is compared with the action data of the standard tennis database by implementing the auxiliary scoring module and given scores. Through the realization of the 3D reconstruction module, the collected tester tennis data is automatically imported into the character model, and the virtual character is driven to realize the function of three-dimensional reconstruction. Methods: The main research methods of this paper are as follows: (1) Documentary Law In the domestic and international important literature databases such as China Knowledge Network, Clarivate Analytics Web of Science, Elsevier Engineering Bibliographic Database, EBSCO SPORT Discus, etc., important research results such as the journal literature, theoretical works, conference reports and dissertations on the application of sports capture in physical education and training are reviewed. It provides theoretical foundation for the comprehensive construction of the overall framework of the study and offer theoretical assistance for the research priorities and difficulties in each specific study. (2) Survey interview Interview with experts and scholars in the fields of physical education and training (web teaching direction), sports human science, physical engineering, and computer application in the form of appointments, conferences, or seminars. Consultation and ask for advice on the difficult issues including biomechanical analysis of tennis movements, human body posture feature extraction, functional module design, database design and so on. In addition, according to the opinions of various experts, determine the specific research plan, adjust the research plan for guarantee the scientific and feasibility of this research. (3) System engineering method The system engineering method can scientifically analyze different factors and functions, and use the modular design technology to divide the system into multiple simple parts. After analyzing the partial and overall conditions, the best solution is developed. The system engineering method is applied in the process of software design of tennis assistant teaching system, thereby reducing the time of software design, speeding up the design process, and improving the quality of software design. (4) Experimental method This paper studies the tennis action data collection through the Kinect equipment that the human body does not need to wear other equipment, and completes the tennis action standard database experiment and the tester\'s tennis action data experiment. Results: The main research outcomes are as follows: The demand analysis of college tennis auxiliary teaching system. In order to better realize the college tennis assistant teaching system, based on the understanding of the concept of demand analysis, the system has taken the following main steps for demand analysis. (1) Clarify the use groups and needs of college tennis assisted instruction systems; (2) Clarify the framework and process of college tennis assisted teaching system; (3) Clarify the technology and development environment adopted by the college tennis assistant teaching system; (4) System testing and feedback. According to the above requirements analysis, it is proposed to determine the overall functional framework of the system as shown in Figure1, which mainly includes the following four functional modules: (1) Video teaching module. The video teaching module is designed to provide students with tennis action video guidance, which is convenient for students to learn by watching videos and to select the content of the tennis technology chapter. (2) Motion acquisition module. The motion collection module includes data acquisition, data filtering and noise reduction processing, and skeleton data storage. The collected standard action data of the tennis teacher is saved after data filtering and noise reduction processing to establish a standard action database; the collected skeleton data of the student is saved after filtering and noise reduction, and waits for the test scoring stage. (3) Auxiliary scoring module. The auxiliary scoring module mainly extracts the angular features of the standard skeleton data collected by the motion collection module. The angle of the test skeleton is extracted, and the DTW algorithm matches the corresponding frame. Then, the velocity feature extraction is performed on the two sets of skeleton data respectively, and the angle feature data extracted before is used to provide the auxiliary evaluation basis for the test module to compare the test skeleton data with the standard skeleton data. Finally, it gives judge, score and suggestions for tennis action learning so as to fulfill the purpose of assisted teaching. (4) 3D reconstruction module. The 3D animation display module mainly assigns the acquired sports files of the student\'s tennis action data to the established character model, drives the virtual characters, reproduces the students’ animations, and makes the tennis movements get rid of the traditional experience-oriented training and increase the fun of training. According to the overall framework design of the system, the college tennis assisted teaching system based on Kinect motion capture is realized by programming. The hardware device consists of a computer (PC: Windows 7 development environment, dual-core 2.13GHz processor, 32-bit operating system) and a Kinect for Windows sensor. The PC is connected to the Kinect sensor via a USB interface. The Kinect sensor is responsible for collecting motion data, and the PC is responsible for real-time analysis and processing of motion data collected by Kinect. Software development tools are C# language, Unity3D, Kinect for Windows SDK, Visual Studio 2012 Express (VS), and .NET Framework 4.0. Conclusions: (1) Through the analysis of the application of motion capture technology to sports training; there are few motion capture systems on the market for in-depth study of tennis motion simulation teaching. Therefore, it is necessary to carry out research on college tennis assisted teaching system based on Kinect motion capture with independent intellectual property rights in China. This paper proposes a college tennis teaching aid training system based on Kinect motion capture, which can make up for the market demand and the shortage of existing similar systems. (2) In motion capture, the human pose motion is divided into a number of corresponding frames as samples. Each frame contains the position and rotation information of all the points of the human body motion at that moment. Therefore, according to the data information, differences in movements between the students and teachers can be judged. At present, mostly used domestic techniques for corresponding frame matching is time-spaced sampling method and so on. This method is simple to implement, but when it comes to slow motion, it will cause data redundancy. Therefore, this project intends to use dynamic time planning DTW algorithm, and the corresponding frame matching of DTW is more accurate than the corresponding frame matching algorithm of equal time interval. (3) In order to shorten the time of software design, speed up the design process, and also improve the quality of software design, this topic proposes to apply the system engineering method in the process of software design of tennis assistant teaching system innovatively. The software design of tennis assisted instruction system is divided into seven stages, such as the proposal of software design tasks, ensuring software feasibility, user requirements for software, software summary design, software detailed design, software coding and testing, and identification and promotion of software. Acknowledgements: This work was supported by 2018 Graduate Education Innovation Foundation of Wuhan Sports University, China (Research on Somatosensory Training System of Tennis based on Kinect Motion Capture). Research on the influence of Tai Chi teaching on physical fitness based on virtual reality technology Shun Wang1, Jie Mao2, Ye Zhang1 1Graduate School, Wuhan Sports University, Wuhan, 430079, China; 2College of Sports Engineering and information technology, Wuhan Sports University, Wuhan, 430079, China Objective: At present, the introduction of virtual reality technology into Tai Chi teaching is a hot spot of studies carried out by education scholars. In this paper, it was studied by changing the teaching mode of Tai Chi so as to improve the teaching effect of Tai Chi. Through the related experimental design, the data obtained from the experiment were analyzed, and different teaching effects obtained by different teaching methods were compared, so as to prove whether the virtual reality technology could promote the teaching effect of Tai Chi and improve the physical quality of Tai Chi exercisers. Methods: (1) Literature data method: Through the retrieval of relevant literatures in databases such as China Knowledge Network, the data on the evaluation indicators of the teaching effect and physical fitness of Tai Chi exercise were reviewed, and the literatures obtained were studied and analyzed, and students’ proficiency and accuracy of Tai Chi exercising were taken as indicators of the teaching effect. Nine body data including chest circumference, waist circumference, hip circumference, upper arm circumference, thigh circumference, body fat percentage, duration of standing with one leg and the eyes closed, step index, and vital capacity were used as indicators of physical fitness. (2) Experimental research method: The experiment selected 20 students without any Tai Chi exercising experience as research subjects, taking the 24 moves of Tai Chi as an example, designing the control group and the experimental group. The control group used the traditional teaching method. A teacher led 10 students to practice Tai Chi exercises; the 10 students in the experimental group performed Tai Chi exercises in the virtual reality scene. The control group and the experimental group were simultaneously practiced for 2hours per day for a period of one month. The teaching effect indexes were the proficiency and accuracy of the Tai Chi movement after the students practiced Tai Chi in different teaching environments, which were scored by Tai Chi experts, with the number 10 being the highest value. The physical fitness indicators were the physical data of the students in different teaching environments, which were measured and evaluated as well. (3) Test scoring method: after the month-long teaching of Tai Chi, according to the requirements of the teaching and examination outline of University Tai Chi Course, combining the scoring criteria of Tai Chi, an assessment team composed of Tai Chi experts conducted an assessment of the technical skills of the experimental group and the control group. (4) Mathematical statistics method: Matlab was used for mathematical statistics analysis. The body data samples collected from the experimental group and the control group were subjected to independent sample t-test. The data were expressed as mean±standard deviation, and P 0.05 was set as significant difference, and P 0.01 as very significant difference. Results: (1) The 20 students who participated in the experiment (without any Tai Chi exercising experience) completed a one-month Tai Chi exercising. In terms of proficiency, 90% of the students in the virtual reality scene teaching mode were more proficient in the 24 moves of Tai Chi than those in the traditional teaching mode. In terms of accuracy, 30% of students in the virtual reality scene teaching mode were more accurate in the 24 moves of Tai Chi than those under the traditional teaching method. (2) The experimental data are as follows: the body data in the control group were chest circumference: 83.17cm ± 3.58cm, waist circumference: 71.02cm ± 7.54cm, hip circumference: 91.12cm ± 5.80cm, upper arm circumference: 25.90cm ± 2.35cm, thigh circumference: 55.31cm±2.39cm, percentage of body fat: 18.31%±4.98%, duration of standing with one leg and the eyes closed: 55.35 s±28.05 s, step index: 57.93±10.55, vital capacity: 3259.5ml±450.3ml; various body data in the experimental group were chest circumference: 83.09cm±2.98cm, waist circumference: 69.17cm±6.35cm, hip circumference: 90.27cm±5.13cm, upper arm circumference: 25.35cm±1.56cm, thigh circumference: 55.15cm±2.37cm, body fat percentage: 17.25%±4.75%, duration of standing with one leg and the eyes closed: 63.25 s±25.37 s, step index: 55.5±13.57, vital capacity: 3350.3ml±430.7ml. Through the comparison of the experimental data, it was found that in addition to the insignificant difference in chest circumference and thigh circumference (p 0.05), the Tai Chi teaching method based on virtual reality technology was obviously different compared with the traditional teaching method in indicators such as waist circumference, hip circumference, upper arm circumference, vital capacity and body fat percentage(P 0.05), and the values of waist circumference, hip circumference, upper arm circumference and body fat percentage in the experimental group were significantly smaller than those in the control group. The value of vital capacity of the experimental group was significantly larger than that of the control group; There were significant differences in the indicators of step index and duration of standing with one leg and the eyes closed (P 0.01), and the values of duration of standing with one leg and the eyes closed in the experimental group were significantly larger than those in the control group while the values of step index in the experimental group were significantly smaller than those of the control group. Conclusion: (1) Through the practice of the 24 moves of Tai Chi, the teaching mode of virtual reality scene showed far higher values in students’ proficiency but slightly lower values in their accuracy compared with the traditional teaching method. It was also found that the teaching method based on virtual reality technology was more popular among students. In the virtual reality scene, students were free from the constraints of the teacher and could practice Tai Chi moves at their own will, which could not only stimulate students’ interest in learning and improve their self-learning ability but also enhance the efficiency of Tai Chi teaching, and from the perspective of training conditions and training costs, training in comfortable indoor venues was also the favored choice of most Tai Chi learners. (2) Tai Chi moves are to strengthen the whole body muscle groups and muscle fiber movements through the arc movements which emphasize the harmony and balance between dynamic and static states of the body, so as to improve the physical quality. Changes in body dimensions and percentage of body fat caused by aerobic exercise can improve body shape. Through the experimental data, it was found that the traditional teaching method could improve the physical quality and posture of students, while the changes of the indicators of students in the virtual reality environment were more obvious. Therefore, the teaching effect of Tai Chi based on virtual reality technology was better. Judging from the influencing factors of teaching effect, (i) For students, the traditional Tai Chi teaching process was boring, and this unchanging teaching mode was not conducive to the development of students’ interest in learning. While the teaching mode based on the virtual reality technology interacted with students in a new teaching environment, and both the sense of immersive experience and the vivid and lifelike images on the screen played a role in stimulating students’ interest in learning. As a result, the negative psychological feelings of resistance and fear in learning, such as the worries of \"unable to remember” and \"unable to learn” of the students, generated due to the complexity of movements, were gradually reduced, and their pleasure and concentration in learning were improved. (ii) The virtual Tai Chi learning environment could break the limitation of time on teaching, and students could practice Tai Chi moves at any time they felt like, which made it possible for students to independently perform Tai Chi moves according to the actions on the display screen, without the presence of a teacher. (iii) The traditional Tai Chi teaching mode relies on the teacher\'s hands-on teaching method, and students practice the moves by watching and imitating the teachers’ standard moves. However, students usually judge whether their movements are standard or not by subjective consciousness when practicing, while a teachers’ energy is limited and it is impossible for him to timely correct each student\'s movements at any time. Whereas, in the virtual environment, there was a model of a Tai Chi coach, which could detect students’ movements and compare them with the standard movements, and feedback the completion status to students, so that they could correct mistakes in time and avoid the occurrence of the phenomenon of \"more practices, more mistakes.” In conclusion, in terms of the teaching effect of Tai Chi moves, the teaching method based on virtual reality technology was superior to the traditional teaching method. Acknowledgment: We thank Qi Luo for helpful discussions and the referees for greatly improving the manuscript. The research was supported by Wuhan Sport University, The East Lake Scholar Foundation. 1Key Laboratory of Adolescent Health Assessment and Exercise Intervention of Ministry of Education, East China Normal University, Shanghai, China; 2Physical Education Department, China University of Geosciences (Wuhan), Wuhan, China; 3School of Physical Education and Health, East China Normal University, Shanghai, China Background: Currently the youth sport is one of the key sports work in China, the national government specially formulated the \"13th Five-Year Plan of Youth Sports” and \"Youth Sports Promotion Plan”. The fast development of Chinese society and economy which improved the physical activity conditions for urban youths, but the number of Chinese youths is very huge, especially in the urban, so the youths have been facing insufficient sports space during a long time, which restricts their participation in physical activity and endangers their physical and mental health. Therefore, it\'s particularly urgent to increase the urban youth sports space effectively with the land resource limit. Methods: After consulting many relevant literatures and interviewing 4 experts, we compiled the questionnaire, and surveyed 962 youth in Shanghai who aged 13-18 to understand their general situation to sports space. The ArcGIS10.2 software is used to analyze the sports space character in Chinese urban, and the Jiang-an district of Wuhan was used as a case. All of the above-mentioned information was used to estimate the usage and development of youth sports space in Chinese cities. Results: It is found that the supply of Chinese urban youth sports space is affected by the youths’ schedule of work and rest, their sports participation interest, family\'s gap between rich and poor, and the distance between their home to the stadiums, etc. And the sports space is relatively insufficient and excessive. Therefore, this study suggests that: (1) formulate a time-division management plan for public stadiums and gymnasiums to explore the service potential of the existing sports space; (2) innovate the development model of urban sports space, and improve the spatial efficiency of sports land, and build more new sports facilities; (3) guide the urban youth to stratified use of the sports space, alleviate the pressure of regional sports space supply; (4) improve the urban youths’ capability to do sports near their home, and bring the non-professional sports space into play positively to realize the diversified growth of urban youth sports space. Conclusions: Chinese urban youth sports space should not be blindly expanded, we should consider the actual character of the urban youths’ physical activity and the situation of the urban sports space supply. So on one hand we should increase youth sports space, and on the other hand we need to improve the service efficiency of the existing sport facilities. Acknowledgements: This work was supported by the National Social Science Foundation of China (Grants No. BLA150061), and The Sixty-first China Postdoctoral Science Foundation (Granted No. 2017M611484). Study on the status of drowning high-risk behaviors and influencing factors among primary and secondary school students Shi Luo1, Yuxuan Ma2 1School of Physical Education, Southwest University, State Key Laboratory of Physical Fitness Evaluation and Sports Function Monitoring, Chongqing 400715, China; 2School of Physical Education, Central China Normal University, Wuhan 430079, Hubei, China Objectives: Students drowning has become the \"number one killer” of abnormal death in China. Downing high-risk behavior is the most important antecedent variable leading to the drowning of primary and secondary school students in China. In order to investigate the current situation of drowning high-risk behaviors and its influencing factors among primary and secondary school students in China. Methods: The questionnaire survey was conducted among 7485 primary and secondary school students in 5 provinces in China. Descriptive statistics and one-way analysis of variance were used to analyze the key variables of drowning high-risk behaviors and their influencing factors. Results: (1) The overall level of water safety education for primary and secondary school students in China is relatively backward. In addition, which is mainly reflected on the low overall scores of water safety knowledge and water safety skills. Besides, there are differences in water safety knowledge between urban and rural areas. The safety knowledge of primary and secondary school students in urban areas is greater than that in rural areas. (2) In terms of water safety skills, rural primary and secondary school students are higher than urban primary and secondary school students. (3) In terms of psychological factors that affect drowning high-risk behaviors, the differences in sensation seeking to exist on gender, urban and rural areas and stages. (4) In terms of environmental factors that affect drowning high-risk behaviors, for the male, rural, and secondary school students, the level of deviant peers was higher. Then, the level of parental behavior control was higher in female, urban and primary school students. Conclusions: Groups in rural China, men, and junior high school students are the targets of important prevention and intervention. Acknowledgements: This work was supported by Ministry of Education Humanities and Social Sciences Research Project of China. (Grants No. 14YJC890017) Empirical study of adopting \"competition-orders” in the interactive teaching activity of SPOC mode Yanwen Wu, Qian Rong College of Physical Science and Technology, China Central Normal University, Wuhan, China Objectives: When it comes to Chinese teaching, the topic of writing teaching is unavoidable. For irrationality in the traditional composition teaching, it is difficult to meet the demand with society development. The specific manifestations are as follows. Firstly, teaching instruction is stereotyped, blindly partial to expression skills yet neglect the depth of thought. Then, lack of lively teaching situation and active multi-dimensional teacher-student interaction, which make the teaching process seem unitary. Finally yet importantly, writing assessment is relatively one-sided by unduly uniformity pursued, with individual differences among students ignored. Methods: To solve these problems, we utilize \"competition-orders” in the interactive teaching activity of SPOC mode with online and offline teaching strategies redesigned. Meanwhile, by improving students’ user model and teaching evaluation system correspondingly, a practical platform is established for students and teachers, on which they can learn and do some interactive teaching activity. The learning platform is including two parts-online and offline. On the one hand, the lack of context and weak sense of engagement can be effectively addressed on the online with rich network resources, which might be the shortcomings in the traditional teaching; on the other hand, the offline part provides efficient teacher-student interaction with solutions of \"competition-orders” matching. Results: To carry out corresponding empirical research, we recruited 120 students to attend teaching experiment and completed the 8-week course. In all, it can be divided into three aspects shown as below: (1) Upon the personalized teaching style, over half of students showed their preferences for teaching materials such as image, text and sound, especially image resource. In addition, approximately 89% learners showed great interest in writing materials during pre-class reading period, such as latest hot political news accounts for 86%. (2) As for the offline teacher-student interaction, number of participants only occupied 9% in the first two weeks. Subsequently, it jumped to 60% in the second two weeks, and fell down to 55% in the third two weeks. Finally, 50% covering became steady in the last two weeks. (3) By collecting the feedback about the intelligent appraisal of online learning, it is shown that over 80% learners’ are satisfied for their improvement both in writing structure and Language fluency. Conclusions: We propose a SPOC-based new mixed teaching mode. It is no doubt that the teaching effect has improved significantly in the SPOC mode than traditional writing class. Due to the boosted interest in writing, learners’ learning ability is enhanced, which promote their personality writing style to be formed. 1Graduate School, Wuhan Sports University, Wuhan, 430079, China; 2College of Sports Engineering and Information Technology, Wuhan Sports University, Wuhan, 430079, China. Objective: Motion sensing technology has become a hot topic of researches among scholars. With the deepening of the related researches, there has appeared many interactive designs of motion sensor in education field. However, there are few experimental studies on the learning effectiveness of motion sensing interaction designs. This paper verifies the effect of motion sensor yoga learning on students’ physical health through experiments. Methods: (1) Literature data method: Search the relevant literatures on the Knowledge Discovery Network Platform with \"Yoga”, \"Effect” and \"Experimental Research” as the key words, and analyze the relevant experimental researches on the effects of yoga learning on physical health, and learn from the relevant indicators selected in previous experiments to provide a basis for the selection of experimental indicators in this paper. Body shape indicators include chest circumference, waist circumference, hip circumference, upper arm circumference, thigh circumference, and body fat percentage. The indicators of physical fitness include grip strength, standing long jump, sitting body flexion, sit-ups, step index, and duration of standing with one leg and the eyes closed. (2) Experimental research: 40 female college students from non-sports majors were selected as research subjects. These college students previously did not participate in regular physical exercise or have any special diseases. They were randomly divided into an experimental group (20 participants in yoga learning under the motion sensing interaction) and a control group (20 participants in traditional yoga learning). There was no significant difference between the two groups in terms of body shape and physical fitness before the experiment. The experimental group and the control group were both given weekly (90minutes each time) yoga learning with the same learning content (same yoga posture, same movement sequence). The control group adopted the traditional yoga practice method, led by a teacher. While the students in the experimental group performed yoga exercises in a motion sensing environment. (3) Mathematical statistics method: SPSS was used to carry out an independent sample t test on physical form, physical fitness and other indicators of the two groups measured in the experiment, and the effect of the two different yoga learning methods was judged through data comparison. Results: Before the experiment, students without significant differences in all indicators were selected for the experiment. After 16weeks of motion sensor yoga learning (experimental group) and traditional yoga learning (control group), there was no significant difference found in chest circumference between the two groups (P 0.05) while significant differences were found on thigh circumference, waist circumference, hip circumference, upper arm circumference, percentage of body fat, grip strength, and standing long jump between the two groups (P 0.05). In specific, the experimental group showed significantly lower values in the indicators of thigh circumference, waist circumference, hip circumference, upper arm circumference and percentage of body fat but obviously greater values of grip strength and standing long jump compared with the control group. There was a significant difference on the indicators of sitting body flexion, sit-ups, step index and duration of standing with one leg and the eyes closed, (P 0.01), whose data values were significantly larger in the experimental group than in the control group. Conclusion: Through experimental data, it was found that although both methods could improve the physical shape and physical quality of college female students, the effect of motion sensor yoga teaching was more comprehensive and significant. The change in physical fitness indicators of yoga learning is mainly achieved by static stretching of the muscles through posture movements, thereby strengthening the performance of the muscles in strength training, and also increasing flexibility. The change in body shape indicators is mainly realized as yoga is an oxygen-tolerant exercise, which has a relatively long exercise time and consumes excess body fat, causing changes in body dimensions and body fat percentage. The main reasons for the differences in the indicators between the two methods are as follows: (1) motion sensing environment stimulated students’ interest in learning and improved learning atmosphere The new style of yoga learning applied the new media to its teaching, and the motion sensing interactions made the originally boring classroom relaxed and enjoyable with its own entertainment nature. In addition, the sense of immersive experience and the vivid and lifelike images on the screen played a role in stimulating students’ interest in learning. As a result, the negative psychological feelings of resistance and fear in learning, such as the worries of \"unable to remember” and \"unable to learn” of the students, generated due to the complexity of movements, were gradually alleviated. By learning in the motion sensing environment, students felt as if they were in a game, and the active learning mentality improved the concentration of learning. (2) Motion sensing environment improved students’ self-directed learning awareness The feedback of the completion of the movements in the motion sensor yoga learning method increased the students’ desire to challenge and stimulated their morale. It was conducive to cultivating students’ self-learning and self-improvement consciousness, forming a student-oriented learning model and highlighting the subjective status of the students. The formation of students’ self-awareness could make them strengthen their self-requirements. Reflected in the learning process of motion sensor yoga, the students could correct their movements in time through the feedback of the completion of the movements in the motion sensor display device, and constantly regulate his or her movements by comparing their own movement trajectory with the demonstration movement trajectory, so that their movements could be more standard. In this way, their exercise intensity and exercise amount were ensured, and the exercise effect was improved. (3) Teachers played a leading role in the traditional teaching environment, while students lacked an accurate cognition The traditional yoga teaching method is mainly centered on the teacher. During the learning process, students mainly imitate by watching the teacher\'s action demonstration, and then check their movements through the mirror. Although the teacher keeps emphasizing the standard action essentials, the students mostly judge the completion of their own actions by subjectivity during the practice and do not have an accurate cognition of the completion of their own actions and are in a passive learning state relying on the teacher\'s supervision. In addition, a teacher\'s energy is limited and he or she can not manage to guide each student\'s movements one by one at any time, so the time cycle of mastering the standard movements is relatively long, making the learning effect not as good as that of learning in motion sensing environment. By introducing the motion sensor into yoga learning, it could reduce the sense of boring and tiredness of learning, stimulate students’ interest and initiative in learning and improve the learning effect significantly. Therefore, by rationally applying some new and high technologies in the field of yoga teaching, we can break the inherent, rigid and unchanging learning pattern, solve the problems in traditional teaching, and realize the better development of the yoga exercises. Acknowledgment: We thank Luo Qi for helpful discussions and the referees for greatly improving the manuscript. The research was supported by Wuhan Sport University, the East Lake Scholar Foundation. 1Key Laboratory of Adolescent Health Assessment and Exercise Intervention of Ministry of Education, East China Normal University, Shanghai, China2Physical Education Department, China University of Geosciences (Wuhan), Wuhan, China; 3School of Physical Education and Health, East China Normal University, Shanghai, China Background: With the development of society,economy and urbanization in China, there are more and more population in the cities. In order to meet the urban youths’ growing needs of sports space for their physical activity, it is imperative to increase the sport facilities for the urban youths during the 13th Five-Years Plan. So firstly it is necessary to clarify the current situation of youth sports space supply, and then formulate some effective development strategies. Methods: This study will take the spots space, primary and secondary schools, and the residential communities in Jiang-an district of Wuhan as the objects, import the longitude and latitude data into ArcGIS10.2 software, and project to vector map. Then analyze the situation of sport facilities, service level and optimization strategies in Jiang-an district with ArcGIS 10.2, and to be visualized through thematic maps. Results: After doing the coordinate distribution analysis, ring buffer analysis and Voronoi analysis with ArcGIS 10.2 which found that: (1) the sports space distribution is related with the development level of economy and society, then formed the sports space\'s skewed distribution which is general in Chinese central urban; (2) limited by land resources, it is difficult to increase swimming pool, basketball court or football field in Chinese central urban, so the effective sports space which meet the youths’ fancy are restricted; (3) there are many areas where is lack of sports spaces or no sport facility, so those youths who are living in such areas are facing difficulty to do sports, and the existing sports space\'s service scope is insufficient; (4)the density of sports space, primary and secondary schools and residential areas which near the Jiang-han Road CBD is higher, and there is a good coupling relationship between sports space and youth living space, but the poor coupling between sports space and youth living space in the fringe urban area leads to a heavy service load of sports space. On the whole this study suggests expanding the service capability of sports space in residential areas, optimizing the layout of sports space in residential communities, purchasing commercial sports space and formulating a service system of sports space for the urban youths. Conclusions: Generally speaking, there is a certain relationship between the layout of urban youth sports space and the level of social and economic development in the central urban, so the development of youth sports space which need to consider the character of youth population distribution. Acknowledgements: This work was supported by the National Social Science Foundation of China (Grants No. BLA150061), and The Sixty-first China Postdoctoral Science Foundation (Granted No. 2017M611484). 1Institute of Education, Changchun Normal University, Changchun, P.R. China; 2College of Computer Science and Technology, Jilin University, Changchun, P.R. China; 3Institute of Media and Communications, Changchun Normal University, Jilin Province, P.R. China Objectives: The behindhand teaching methods, the insufficient teaching corpses, and the inadequate original 3D anatomical models have always been the main problems of the teaching of anatomy. With the advanced 3D animation technology, we select the human skull to construct a suitable model for medical 3D anatomical data, thus fully acquiring the 3D data as well as the information about the shape, physique and structure of human body. Methods: Physical photographs and anatomical teaching sampling are combined. Before modeling, the sketches based on the combination of physical photographs and references are drawn to ensure the applicability of the 3D model. The DEFMATION technology in the model construction can reduce the overall number of point-cloud of the model, while its model topology conforms to the ring topology in film and television animation, thus generating a low-precision model with less occupation of computer resources. Results: The product can basically meet the needs on details and product function positioning as seen from the final effect. In constructing the model for a large quantity of detail processing, the subdivision method is adopted, which, as an important research in computer-aided geometric design and graphics, has become one of the most popular research in the field of geometric modeling in recent years, meanwhile, as well as a most ideal method. Conclusions: High-precision 3D skull model can effectively motivate students to learn basic medical knowledge whenever and wherever possible and greatly increase learning efficiency, as well as stimulate learning interest. At the same time, it can transform anatomical education into a basic public welfare for people to achieve a win-winsituation with social and economic benefits. Acknowledgements: This work was financially supported by the 13th Five Years Programs for Social Science Research of the Education Department of Jilin Province (JJKH20181193SK), and the Higher Education Foundation of Changchun Normal University (No. [2017]11). Application of learning community platform for college students’ innovation education Yanwen Wu, Yuezhang Wang College of Physical Science and Technology, China Central Normal University, Wuhan, China Objectives: The core of innovative education is training the cultivation of innovative abilities, which requires a lot of practical activities with high demand on quality teams. Based on this, the paper designed and developed a learning community platform for the innovative education of college students. This study also evaluated the improvement and the quality of innovative educational practices to test the applicability of this platform. Methods: In this study, the learning community platform was designed and developed based on WeChat Mini app. This platform system is composed of four layers: the interaction layer, the application layer, the interface layer, and the infrastructure layer. The main functions they implement are as follows. (1) According to the application of a distributed file system, the infrastructure layer organizes the shared folders together from different servers. (2) Based on the programming development environment, the interface layer customizes API and Web Server modules to transmit data. (3) The application layer consists of three large modules: user, learning, and personalized recommendation. The personalized recommendation module mines the user\'s personalized behavior and personal information, provides personalized recommendation services, and helps users form a learning community. The learning module is a follow-up service module that integrates a large number of teaching resources and can provide targeted teaching and counseling services for users. After the learning community platform completed, we carried out empirical study for a semester, then the data were obtained both from the platform and from the questionnaire survey to evaluate the applicability in the innovation education among college students. Result: Participants in this study are undergraduate students at Central China Normal University, who are going to participate in the National Mathematical Modeling Competition for College Students. The empirical research shows:(a) With the assistance of this platform, the formation rate of the digital model training team in the semester increased by 14.43% compared with the average rate of the past three years, and the training completion rate increased by 7.21%. (b)The platform mainly contributed to the exchange of ideas, cross-faculty exchanges, and team collaboration. (c)96.77% users thought that the platform should be oriented towards more innovative educational practices. Conclusions: The Learning Community Platform provides a place for a large number of individual students to learn and exchange ideas, which breaks the constraints of the personal communication circle and enhances the formation rate of learning community. In this way, the platform promotes the sharing and exchanging of a large number of high-quality ideas, and application these ideas into practice, which has played a positive role in the innovative education of college students. We suggest that other universities should embark on the relevant construction of the learning community platform as an efficient channel for creative communication among college students. International College for Chinese Studies, Nanjing Normal University, Nanjing 210097 China Objectives: The vocabulary teaching is the core of the teaching Chinese as a foreign language. The computer corpus can provide the frequency statistics of the vocabulary, and the probability statistics based on the real texts can optimize the specific contents of the vocabulary teaching. The natural corpus arranged in the computer corpus is derived from the real verbal communication that people convey information. For the second language teaching, it can ensure the accuracy of the vocabulary description and the timeliness of the creation of the lexicon and vocabulary. The quantitative frequency statistics is of great theoretical and practical significance to the selection of the syllabus for teaching Chinese as a foreign language and the hierarchical representation of different vocabulary units in the vocabulary system. A large number of the practical teaching have proved that the data-driven vocabulary teaching based on the computer corpus resources can greatly improve the efficiency of the vocabulary teaching and enhance the effect of the students’ vocabulary acquisition. However, the promotion of this teaching method in teaching Chinese as a foreign language needs further technical support from the relevant government departments and the perfection of the professional knowledge of the teachers for teaching Chinese as a foreign language. Methods: Literature review. Results: The data-driven vocabulary teaching based on the computer corpus should first look for a corpus with strong applicability. At present, the parallel corpus for teaching Chinese as a foreign language has not been established, and the existing parallel corpus is mainly English-Chinese, which cannot meet the needs of the non-native English learners, and its scope of application is limited. The database is mainly targeted at the native speakers of English. Xiao Zhonghua and Dai Guangrong (2010:190-200) pointed out that \"In order to bridge the gap between the corpus and the language teaching, the first step is to create a corpus for our teaching, and from the design of the corpus to the contents of our teaching, all need to meet the curriculum requirements. Only in this way can the corpus-based learning activities become an integral part of our language teaching.” Conclusions: The data-driven learning is a double-sided revolution combining the technologies with the teaching concept for TCFL in the information age. As the main body of our classroom, teachers are the key to the teaching effect. The data-driven vocabulary teaching can update the concept of teaching Chinese as a foreign language in time, make full use of the new computer education technologies such as the multimedia, optimize the curriculums of the vocabulary teaching, and improve the quality of the vocabulary teaching. At the same time, vocabulary, as the most rapidly changing component of the language elements, always reflects the latest situation in the political, economic, scientific and technological fields. By collecting or establishing the computer corpus, teachers can constantly update the computer knowledge, which not only enriches their teaching resources, teaching methods and teaching means, but is also conducive to providing a development direction for the application of the computer corpora and other related fields. Analysis of the service transformation of the textile industry and its value based on the e-commerce Pei Chang Wuhan Textile University, Wuhan, Hubei, 430073 China Objectives: Since 2013, China\'s economy has entered a new normal. The most remarkable feature is that the growth rate of the manufacturing industry has declined seriously, which also affects the textile industry. The competition in the domestic textile market is extremely fierce, and the raw material costs continue to rise, and especially labor costs. Some textile enterprises have a serious problem of overcapacity. In the Internet age, the textile industry has also experienced a new development mode with the help of the easterly wind of the Internet. The Internet has the remarkable characteristics of high efficiency, convenience, wide participation and rapid dissemination. The traditional textile enterprises can use the Internet to realize the transformation and upgrading. It can innovate the new business model of the textile enterprises and enhance the core competitiveness of the textile enterprises. Under such a development background, the textile enterprises should make good use of the \"Internet +”, seize the development opportunities, break the restrictions of the traditional development mode of thinking, improve their service levels of enterprises, and gain the market foothold. At the same time, it is also the key to enhance the competitiveness of the entire textile industry chain and realize the upgrading of the textile industry. However, the current reality is that the textile enterprises have a lot of space for the scope and depth of e-commerce. Although some textile enterprises also recognize the values of the e-commerce and are trying to implement the e-commerce development model, the vast majority of the textile enterprises are still waiting and doubting the e-commerce mode. The promotion of the sub-business mode in the textile enterprises still has a long way to go. This paper studies and analyzes the implementation mode of the e-commerce in the textile industry, and puts forward some solutions, hoping to help enhance the overall competitiveness of the textile industry. Methods: Literature review. Results: There are some problems in the application of the e-commerce in the textile enterprises, respectively are under the Internet mode, the product infringement is more serious, and poor experience of the network platform. The effective implementation of the e-commerce mode in the textile enterprises include enhancing the enterprises’ innovative consciousness, speeding up the personnel training, providing the adequate legal protection, and strengthening the platform construction. In addition, it is very important to strengthen the basic services to promote the continuous improvement of the e-commerce mode in the textile enterprises. Conclusions: There are some differences in the regionality and the productivity of the textile enterprises in China, which also leads to the gradual and arduous implementation of the e-commerce model in the textile enterprises. Even so, the development trend of the application of the e-commerce mode has been irreversible. At present, the brand enterprises in China\'s textile enterprises have been successfully using the e-commerce mode, and have accumulated a lot of useful experience. We believe that, driven by the benchmarking enterprises in the industry, China\'s textile enterprises will fully implement the e-commerce mode, and the e-commerce mode in the textile enterprises will have better and better prospects for its application. Analysis on the management reform of cross-regional tourism integration and development Li Feng Wuhai Vocational and Technical College, Wuhai City, Inner Mongolia, 016000 Objectives: Our country is rich in tourism resources, and many tourist resorts are very famous in the world. Compared with other countries in the world, our country\'s tourism resources are very rich. China\'s tourism industry has been developing rapidly for about 20years. After such a long period of rapid development, the scale of the tourism industry has expanded rapidly, the corresponding tourism products have become more abundant, and the competition in the tourism market has become more intense. Under this general environment, the tourism industry around the country has strengthened its understanding of the management concept of integrated development, and through this management mode, the overall image of regional tourism has been shaped, so as to realize the reform of the tourism industry and promote the effectiveness of competition. However, in the actual process of integrated development, there are still some problems affecting reform and development. Therefore, it is necessary to study the relevant theoretical basis and continuously explore the regional tourism integration and development on this basis, summarize the influencing factors in the actual integration process, follow the corresponding guiding principles, and finally effectively avoid some industrial risks existing in the tourism development process through the specific measures of the integration contents, so as to promote the regional tourism resources to obtain very high utilization benefits. This paper starts with the core elements of the integration of domestic cross-regional tourism, then discusses the actual development status of the integration and development of cross-regional tourism, and further analyzes the specific contents and corresponding measures of the integration and development of cross-regional tourism through the relevant basic principles to be followed. It is hoped that the views discussed in this paper can promote the reform of the management mode of domestic cross-regional tourism integration and development, and relevant personnel of cross-regional tourism integration and development can reform the management of domestic cross-regional tourism integration and development on the basis of relevant core elements, so as to promote the healthy and long-term development of domestic tourism. Methods: Based on the theoretical research on the integration and development of cross-regional tourism. Results: The developer behavior, tourist behavior and the integrated development reform on the behavior of scenic area operators have different impacts on integrated development reform. In order to promote cross-regional tourism integration and development management reform, we can take the tourism market as the basis for the integration of tourism products, take space as the basis for the integrated development of tourism regions, and take industry as the basis of tourism culture integration. Conclusions: To sum up, the integration and development of cross-regional tourism is a measure that conforms to the development of the times. It is necessary to reform its management, but it is also a very complicated system project. In the process of integration and development of cross-regional tourism, its driving force is composed of internal and external driving forces. Its internal driving force is based on the development of the region and its surroundings where the cross-regional tourism is located, while the external driving force is based on the government\'s promotion, its complementary advantages and the construction of the ecological environment. Through the internal and external driving forces, it builds its driving force for the reform of cross-regional tourism development management and actively promotes the healthy development of cross-regional tourism integration and development. 1School of Water Conservancy and Environment, University of Jinan, Jinan, 250022, Shandong, China; 2Shandong Provincial Engineering Technology Research Center for Groundwater Numerical Simulation and Contamination Control, Jinan 250022, Shandong, China; 3Water Resources Research Institute of Shandong Province, Jinan 250013, Shandong, China Objectives: Water is the source of life and the basic resource for the economic and social development. As an essential factor in the development of the human society, the development and utilization of water plays a very important role. The development and utilization of the water resources in the Yangtze River Basin is an important guarantee for the sustainable development of the Chinese nation and the construction of the ecological civilization. At present, the construction of the ecological civilization has a very high requirement, and the development of the Yangtze River economic zone. This paper will study the distribution characteristics of the water resources in the Yangtze River Basin. At the same time, the development and utilization status of the water resources in the Yangtze River Basin and the protection progress of the water resources in the Yangtze River Basin will be analyzed. And the key ideas and the systematic ideas of the development and protection of the Yangtze River water resources in the long-term future is studied, to provide important policy support for the realization of \"development in protection and protection in development” in the Yangtze River Basin in the future. Methods: Literature review. Results: As the basic work of the water resource development in the Yangtze River Basin, the research on the water resources needs to be strengthened in the following aspects: water resources bulletin and water resources fee projects; water resources demonstration and continuous water intake assessment; study of the prediction and operation of the cascade reservoirs; demonstration of the sewage outlets in rivers; water quality monitoring and management; supervision and management of the water users. Attention should be paid to the following four aspects in the utilization of the water resources in the Yangtze River Basin in the future: laws of changes of the water resources under the influence of the human activities; sponge city and urban hydrology; small and medium flood dispatching and rainwater resources utilization; water resources under the climate change. Conclusions: The study of the distribution of the water resources in the Yangtze River Basin will make an important contribution to the economic and social development in the basin. Just as the idea to promote the great protection and green development of the Yangtze River put forward by the Party and the Central Committee, we should always establish and adhere to the concept of \"ecology first” in the study of the distribution of the water resources in the Yangtze River Basin, adhere to the research orientation of the people-centered water resources distribution, and promote the safety and security of the water resources in the Yangtze River Basin. Research on modern Chinese history in the era of big data: opportunities and challenges Haifang Zhang, Wei Wang Nanjing Normal University, Nanjing, Jiangsu, 210000 China Keywords: Big data; Opportunities and challenges; Research Background: The big data brings challenges and opportunities to the study of the modern Chinese history. The big data not only challenges the researchers in collecting the historical data, their thinking mode and research levels, but also makes it very convenient to retrieve and consult the data. It opens up the horizon of the historical research and expands the depth and breadth of the information researchers possess. Methods: Information research (data processing); Association analysis method. Results: We should play the role of the big data in the study of the modern Chinese history from three aspects. Firstly, we should further improve the database, strengthen the data processing of the existing paper data and information, and realize the interconnection and interoperability of the related databases with an open and mutually beneficial attitude. We use ADO.NET\'s dynamic data binding technology to link the data of modern history research with ADO.NET and database. When developing the database, the SqlConnection object and name are used, and the SqlConnection object can be used to delete, update, query and other related commands for each history, and the SqlDataReader returns the record for the user to query. In order to facilitate the sequential reading of database material content, the records returned from the SqlDataReader are designed to be executed by the data stream. That is, the data stream obtained from the database can only be read out in order, and the user query record can be easily and quickly responded. If the user wants to manipulate the record under certain conditions, the best way to do this is to use a DataSet. Secondly, we must create the quantitative research paradigm. In the course of the study, we choose the data in a long period of time and with a large scale, focusing on the macro-scope and integrity of the research. In the process of Chinese modern history data research, the most basic data structure of all data is a data structure stored in a linked list, combined with traversal and multiple keyword traversal methods. The main types are: double-linked list data structure and single-linked list data. structure. Among them, in the doubly linked list data structure, there are two chains with different directions, that is, in each node, in addition to the next domain storing the subsequent node address, a pointer field is added, which points to its direct predecessor. In this data structure, it should be noted that the double-linked list is determined by the head pointer. Thirdly, the association analysis method is introduced, to mine the relativity of things existing in a large number of the data, so as to find out the main contradictions and the general laws of the development of things. Currently in China, the goal of Web Miner, a data mining technology for modern history, is to implement a human-computer-based Web-based data mining system. The object of mining is a certain kind of thing specified on the Web. And text format learning, extract useful data from the data warehouse, and then perform association rule mining and data classification in the data warehouse. The whole system is mainly composed of five parts: Web Obtainer, Web Resource Acquisition Module, Web Filter, Web Mining, Miner, Data Warehouse and a domain knowledge base for mining. Different from the information in the dynamic database of ADO.NET, the information on the Web is characterized by disorder, unstructured, dynamic, and diverse. Sometimes there may be redundancy, inconsistency or even contradiction between data sources, so Web-based the data mining work in China\'s modern history is very difficult. The system and work described in this paper are both preliminary and groundbreaking. Conclusions: Firstly, The DataSet supports the disconnection of data, and can help manage the data in the memory. It can extract data from the data stream in order and in an unordered manner, greatly improving the efficiency of the program. In addition, using the DataGrid control DataSet class through the ADO.NET database, you can bind the modern history research database without code. Secondly, by taking the lead node, some operations are more convenient in the entire double-linked list system. By combining the head node and the tail node, a bidirectional circular linked list data structure is constructed. Because the current Chinese modern history data research only pays attention to the logical order between the nodes, and does not pay attention to the position between the nodes, the arrow application is used to specifically represent the pointers in the chain, linear tables, such as bat, cat, etc. Linked list type. Thirdly, the application of Web Miner data mining technology to the WWW network, especially the combination of search engine technology on the Web, will undoubtedly bring huge benefits and bright prospects to the discovery, collection, query and utilization of Chinese modern history information on the network. Research on the development of smart medical APP for cervical and lumbar vertebrae soothing based on the theory of preventive treatment of disease Senmeng Gao1, Rui Zhou1, Peilu Tang1, Chunjia Wang1, Lingfeng Fei2 1College of Art and Design, Xihua University, Chengdu, China; 2Department of Digital Art, Chengdu Neusoft University, Chengdu, China Objectives: The concept of preventive treatment of disease for medical cervical and lumbar sub-health populations has positive significance for health management, and is particularly applicable to urban life groups who work at long-term desks. Combining the medical theory of treatment of diseases with current smart hardware devices, the APP that is suitable for sub-healthy working young people will be developed to specifically relieve cervical and lumbar lesions. Methods: Through questionnaires and in-depth user interviews, the typical working life forms of the target treatment population were excavated, and the smart wearable medical equipment was analyzed for its intervention in the normalized treatment of cervical and lumbar spine. In combination with competitive product analysis of mainstream medical APP in the market, smart medical care was constructed. Cervical lumbar spondylosis APP function framework and development program. Results: The project demand study found that more than 70% of the young Chinese working people ignored the initial cervical and lumbar spine lesions, which was suitable for intervention with the medical treatment of \"preventive treatment of disease”. Two modules of active medical care and passive medical care were constructed in the cervical and lumbar medical app. The active medical module takes advantage of the mobile terminal in the APP function to achieve health monitoring and treatment reminders. The passive medical module uses smart wearable devices to achieve a normalized treatment for patients with mild conditions. Conclusions: The idea of \" preventive treatment of disease \" can be an important strategy for the prevention and treatment of cervical and lumbar spondylosis. For the sub-health population medical plan, it can fully combine artificial intelligence, intelligent hardware technology and mobile terminal APP, so as to realize the establishment of personal health management concept and daily continuous treatment, and fully practice the \" preventive treatment of disease \". It has three medical aspects: prevention before disease onset,prevention of disease from exacerbating and prevention of recurrence. Acknowledgements: Supported by the 2018 Innovation Fund of Postgraduate of Xihua University (Grants No. ycjj2018163); The Research Project of the Key Research Base of Philosophy and Social Sciences \"Sichuan Cuisine Research and Development Centre” (Grants No. CC18W08). Medical physiotherapy product design research for the delayed onset muscle soreness symptom Yunqi Zhang1, Yiming Huang2, Xun Luo2, Jinfeng Lei3 1College of Art and Design, Xihua University, Chengdu, China; 2Department of Culture Industry, Chengdu Museum, Chengdu, China; 3College of Arts, Guilin University of Technology, Chengdu, China Objectives: The User-centered Design (UCD) design philosophy and process are involved in the development of medical equipment products, which specifically explores the development of Delayed Onset Muscle Soreness (DOMS) physiotherapy products after exercising and promotes the sustainability of athletic behavior of the target users. Methods: For sub-healthy people lacking exercise and sports, through questionnaires and user logs, we focus on understanding and digging into the target audience needs of physiotherapy products. A physiotherapy project was constructed through expert interviews to relieve the Delayed Onset Muscle Soreness after exercising. From user-experience perspective, new technologies such as the Internet of Things and human interaction are used as a consistent point to form Online to Offline (OTO) and materialized informatization development for alleviating Delayed Onset Muscle Soreness physiotherapy products. Results: A wearable device is used as a physical product carrier for physiotherapy project, which equipped with low temperature cold therapy and Chinese massage treatment. The reconstruction is based on the information products of mobile internet information technology and intelligentized \"Internet of things”, through interaction design, which forms the relieving Delayed Onset Muscle Soreness remote physiotherapy service system. Conclusions: Based on the innovative user experience design, it provides many new ideas for the development of physiotherapy medical equipment, and offers the basis and reference for remote precision medical services. It will enhance patient\'s compliance through online to offline (OTO) system development of intelligent physiotherapy product. Small medical products enter Chinese households, and medical equipment is transformed into small household appliances for improve the life quality of sub-healthy and elderly people. Acknowledgements: Supported by the Research Project of Sichuan Provincial Education Department Key Research Base of Philosophy and Social Sciences \"Industrial Design Center” (Grants No. GYSJ17-018); Guangxi Young and Middle-aged Teachers’ Basic Ability Upgrading Project (Grants No. 2018KY0229). Objectives: Recently, diversified operations with different styles and different degrees were developed successively but success rates are few. In the diversification management, such factors as rational utilization of resources sharing, diversification of risks in resource combination, and improvement of enterprise market power have a positive impact on enterprise performance. the opposite negative factors include cross-industry investment risk, resource dispersion, agency problem, cross-subsidy, and management cost increase are not to be neglected. It is important to analyze the correlation between diversified operation and health investment. Methods: Choosing diversified operation sufficiently and reasonably is very necessary for sustainable development of company. Depending on this act, the company will come back to its main operating project and abandon business which is irrespective with major works, thus the company will have competitive in the increasingly competitive society. That is the reason why SASAC introduces economic value added (EVA) to enterprises. Meanwhile, application of enterprise diversification will disperse effective resource and the major works cannot develop better, which decrease the profit margin of the company. Results: According to correlation analysis, negative correlations were found between enterprise diversification and over-investment. That means company could through adopt the strategy of diversification to decrease the possibility of over-investment. In addition, several factors including core business value, free cash flow, company scale, asset-liability ratio will restrain over-investment. Through related research, we can know that growth opportunities, free cash flow, enterprise scale, management expenses are positively and significantly correlated with the over-investment. Conclusions: Research finds that the diversification degree and over-investment are presented as negative correlations. That means the higher of the diversification degree of the company, the more restrictive of behavior of the over-investment. On the other hand, examination of applied economic added value will realize a positive correlation between diversification degree and over-investment. Moreover, application of economic value examination will realized a positive correlation adjustment, which remit conflicts between company and agency and restrain over-investment, thus improve healthy development of company. Acknowledgements: This work was supported by the Science and Technology Plan Projects of Shaanxi Provincial Department of Education (Grants No. 18JK0935). 1College of Information and Communication Engineering, Dalian Nationalities University, China; 2College of Physics and Materials Engineering, Dalian Nationalities University, China Background: Medical natural text mining is a very tempting and technically challenging area. On the one hand, data scientists working in the field face the strong demand and desire of medical scientists for the analysis of medical text data. On the other hand, they have to face the complexity of medical text data and the difficulties of many mining, as well as the bottleneck of current related technologies. In addition, since the medical text data is necessarily based on a specific natural language environment, the medical text data of different natural language environments have different characteristics, and the related technologies of medical natural text mining have similarities and differences. In the academic literature, we can find a large number of related papers on medical natural text mining, many of which are based on theoretical and experimental aspects. The truly mature technologies and methods that can be effectively applied in practical work are actually limited. This reflects the complexity of medical text data and brings many challenges to researchers. Methods: We have developed a medical text data mining system, which is based on large-scale medical clinical data. It uses various techniques of data mining technology such as feature extraction, text classification, text clustering, etc. Combined, it is applied to descriptive clinical text data and medical image reports and other text materials to explore a large number of laws with medical value. The text feature selection algorithm adopts the feature evaluation function method, uses the document frequency method of the minimum word frequency threshold to select features, and uses the information gain feature evaluation function to perform the operation. Experiments show that the proportion of noise features in feature set can be effectively reduced, and it is found that the feature set obtained by the evaluation function tends to be consistent with the increase of threshold value. This analysis system with descriptive group medical data as the research object can be mined. The diagnostic description features of a disease in medical natural texts, as well as the potential effects of various examination indicators and diseases, can also dig the diagnosis mode of a disease in imaging examination, and explore the law of the simultaneous occurrence of different symptoms. The text data mining system can also perform text classification analysis and text cluster analysis on textual information such as medical natural texts and inpatient medical records, and has achieved good results. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 61771092). Objectives: With the development of era, the educational concept has been updated continuously. The real purpose of education is cultivating learners to designers of their own knowledge structures, thus learners will be good at learning. Ability training is not only imparting knowledge through teaching, but also improving cognitive activity and cultivating innovative abilities. The purpose of the medical college education is to cultivate talents with technically complete technologies. High hands-on operations, independent thinking, and the ability to integrate multi-disciplinary knowledge are necessary for medicine students, so medicine comprehensive experimental reform is imminent. Methods: Comprehensive and exploratory experiments are also called designed experiments or simulated scientific researches. For unknown issues, students driven by exploratory comprehensive experiment to use scientific way of thinking to design the experimental ideas. The exploratory experiment can cultivate the ability of students to apply comprehensive knowledge of various disciplines for comprehensive and innovative thinking. Results: Students who take exploratory experiments can not only enhancing understanding of textbook theoretical knowledge, but also improving experimental design. Students will forwardly scanning in database and consulting professionals, and their learning become more intensively. The results show that the comprehensive analysis ability of students has been improved, and consciousness of innovation also has been cultivated. This will save a large number of manpower and material resources, avoid duplication of expenditure, and realize resource sharing. Conclusions: Experiment course teaching is irreplaceable in the area of cultivating theoretical knowledge application, problem analysis and solving, innovative awareness for students. Traditional education mode focused on teachers, and students were always in a passive position. The comprehensive experiment broke the traditional teaching model which centered on the class and centered on the books. Comprehensive experiment established the concept of the main body of the students’ teaching activities, which fully mobilized their enthusiasm and initiative. By reforming the structure of traditional experimental courses, a comprehensive experimental teaching has been formed, enabling students to learn more knowledge and skills in a limited period of time. It has also strengthened practical training and cultivated self-educational innovation and practical ability. Objectives: The existing hospital gown sets a separate velcro or zipper on the sleeve. The design of separate zipper will make the clothes drafty, heat preservation effect will be poor in the cold season, and the patient will easily catch cold when taking blood. setting the magic tape alone is easy to deform the sleeve, affect the integrity and aesthetic of hospital gowns. So, design a new and functional hospital gown is of great significance. Methods: To Achieve these objectives, a new functional hospital gown is designed. The hospital gown of the elbow socket was hollowed out and covered with patch. The patch was connected with the sleeve through a u-zipper. The outer sheet is set in the sleeve, the outer sheet was covered in the U zipper and the patch, the outer sheet is connected to the sleeve by a magic tape. The drawstring was set in the top of the outer sheet. Outside of the u-zipper, there were two levels fabric with tourniquet inside. The fabric was seamed with the sleeve. One side of the Velcro was set around the u-zipper, and the other side was set on the inside of the outlay and wider than the sleeve. Results: The implementation part of the design was in the elbow socket of the sleeve. The connection was applied by the combination of u-zipper and Velcro. A patch is fixed by a zipper, the covered by a outer sheet. The outer sheet was connected by a magic tape, and it is helpful to make sure the sleeves are not distorted, to keep the integrity and aesthetic of hospital gowns. The drawstring was set in the top of the outer sheet, and it is convenient to open the magic paste. The tourniquet was set in the top of the chelidon and will brings great convenience to the medical staff and improved the efficiency of blood collection. Conclusions: The design is to provide a hospital gown with the function of easy blood collection and warmth maintainment. The fabrics are all made of viscose fiber to avoid enduring cold and bring convenience to the medical staff. Acknowledgements: This work was supported by Shaanxi Province Education Science Planning Topic \"Empirical Research on Teaching Reform of the Course Fundamentals of Computer Application Based on Working Process Systematization in the Background of Transformation and Development” under Grant No. SGH17H461. It is also supported by the Natural Science Foundation of Xi\'an Peihua university \"Research on dynamic migration algorithm of OpenFlow switches based on attractor selection in cloud environment” under Grant No. PHKT18062. Development and application of hospital chronic disease real-time monitoring system in chronic disease management Yubing Xu Infection Management and Disease Control Department of Hainan Branch of PLA General Hospital, Sanya, Hainan, China Objectives: The monitoring and management of chronic diseases in the hospital, to overcome the physicians and staff management report manually or by a doctor is allowed in the computer information system, low efficiency, omission, bottleneck, of monitoring work efficiency and improve the quality of statistics and research hospital real-time monitoring system for chronic disease, chronic disease hospital patients can be automatically screening, statistics and reporting capabilities. Reach as long as the doctor made a chronic diseases related to diagnosis, can complete to collected the basic information of the case, check the result and so on all the relevant information, so as to realize new cases of monitoring, information collection and exchange, data management, statistical analysis and quality control, and with the country\'s chronic disease monitoring report system information docking. Methods: Based on the hospital information system, using Client/Server architecture, based on Oracle database, and the existing hospital information management system (HIS), laboratory management system (LIS), and check the system to achieve seamless connection, With the disease diagnosis as the core information collection \"trigger point” and the real-time operation as the collection method, the related information of hospitalized and out-patient patients can be automatically collected without omission according to the relevant diagnosis of doctors. Results: The system can automatically collect medical information of patients with chronic diseases in designated areas. Its features and innovations include: first, the system can automatically collect information in real time. 2 it is. The monitoring scope of patients with chronic diseases in the outpatient and inpatient areas is all covered, and no missing reports and repeated reports are allowed. Thirdly, the system is scalable. According to the monitoring requirements, the monitoring scope of disease and the chronic events in the monitoring period can be increased at any time, so that new cases can be easily mastered and distinguished. Fourth, the exported information can be directly imported into the national monitoring system in batches through rearrangement. Conclusions: The real-time monitoring system of chronic diseases in the hospital enables timely reporting of clinical and in-patient chronic disease events in the hospital, with accurate and comprehensive information, no omission and full coverage, providing comprehensive and reliable data for the CDC, facilitating the follow-up and health guidance for patients with chronic diseases. 1Department of Mathematics and Statistics, College of Science, Huazhong Agricultural University, Wuhan, China; 2Department of Economics, College of Economics & Management, Huazhong Agricultural University, Wuhan, China; 3Department of Mathematics, College of Science, Louisiana State University, Baton Rouge, Louisiana (U.S.) Background: The aged population in most countries is rapidly increasing in the 21st century due to fertility decline and improvements in medical care. Then health issues of elderly cause social attention recently. We analyze health status of elderly people and its influencing factors, then build a comprehensive health evaluation index system. The aim is to provide advice to medical workers about evaluating the physical and mental health of the elderly. Methods: Analytic Hierarchy Process (AHP) method is used to construct the model frame of comprehensive health index system for elderly based on demand acquisition, analysis, design and implementation. Results: The comprehensive health index system for elderly considers five aspects of the second index level: physical function, self-care function, mental health, cognitive ability and social relations. And thirteen indexes are selected for the third level, including organ activity function and so on. The data is randomly extracted by the ratio method and abnormal data is eliminated by RANSAC method, then the data is normalized. According to the difference between the universal level of social data and the international standard, we design weights of the two index levels. A data-oriented index importance evaluation matrix is constructed, and the consistency test is carried out. Finally, Analytic Hierarchy Process (AHP) method is utilized to evaluate the health conditions of older adults. Conclusions: AHP method is feasible in the establishment of comprehensive health index system for elderly people and provides suggestions for the development of comprehensive health evaluation methods for elderly. At the same time, the aspects in which older people are weak can be shown based on the model. These parts need to be paid more attention. And we can also continue to analyze the dynamic changes of general health conditions of elderly according to this model as time changes. Acknowledgements: This work was supported by the postgraduate curriculum construction of Huazhong Agricultural University \"Longitudinal data analysis with SAS application” (Grants No. 2016KC14). Study on the risk of knee cartilage injury in the landing moment of freestyle skiing athletes of aerial skills Yanming Fu1,2, Tianbiao Yu1 1School of Mechanical Engineering and Automation, Northeastern University, Shenyang, Liaoning, China 110819; 2Laboratory Management Center, Shenyang Sport University, Shenyang, Liaoning, China 110102 Purpose: Knee injury of China\'s freestyle skiing athletes of aerial skills project accounts for up to 80% of the total injuries, and the most common injury is knee cartilage injury. This study is to reveal the knee cartilage injury mechanism and cartilage damage condition of athletes engaging in this project. Method: This study uses kinematics analysis to analyze the freestyle skiing aerial skill, two-circle movements of bLL and bLT, and obtains relevant kinematic indicators and joint angles. The DICOM data of volunteers are obtained by CT and MR detection and the data are imported into Mimics software to establish knee joint model of IGES format. The finite element analysis of the IGES model is performed by Abaqus finite element software to obtain the stress condition of knee cartilage and meniscus. Results: The obtained angles of the right and left knee joint in the landing moment are αL=158.6±13.5(°) and αR=163.0±11.1(°), and the obtained angles of the right and left knee joint in the landing stability are α′L=138.6±13.5(°) and α′R=140.3±16.7(°), so that the study is conducted on the benchmark of 140˚ of knee joint. The maximum stress F=4544.63(N) of the tibial plateau in the landing moment is calculated by the take-off speed, the height of arch and the buffer time. It is estimated that the tibial plateau will bear 6.76 times body weight. The finite element analysis results show that the lower part of femoral cartilage\'s internal condyle and the back end of femoral cartilage\'s external condyle have stress concentration, and the middle part of the meniscus medialis and the middle and post part of the meniscus lateralis also have stress concentration, while the stress distribution of the medial and lateral tibial cartilage is relatively uniform except the center position. Conclusion: The impact to the knee joint cartilage is nearly 7 times the body weight in the landing moment of the two-circle movement. Although there is stress concentration in the femoral cartilage, meniscus and tibial cartilage, the bLL and bLT movements in this study have no transverse axis rotation, so that the forces on both legs are basically balanced. It is worth noting that the accumulated training will cause the cartilage having stress concentration injured, resulting in fatigue damage of the knee joint cartilage. This change in microstructure will result in unpredictable new stress concentrations. Fund project Key Special Project of National Key Research and Development Program \"Technical Winter Olympics” (2018YFF0300502). 1College of Computer Sciences and Technology, Changchun University, Changchun, China; 2College of Optical and Electronical Information, Changchun University of Science and Technology, Changchun, China Objectives: From grid computing to cloud computing, a centralized big data storage and processing center has been formed. Problems such as high energy consumption, high traffic and high delay are brought, and there is also the risk of privacy data leakage. With the advent of a large number of wearable intelligent health care terminal devices, a big data on health care needs to be processed in real time will be generated. The processing with cloud computing model alone can\'t meet the needs of users, while the edge computing model will solve related problems. Methods: The wearable intelligent health terminal equipment with CPU or GPU can realize real-time storing and processing of health data collected by various sensors. It is adopted to build a three-level architectural model for the health care intelligent terminal. The highest layer is cloud server cluster that analyzes big data to generate treatment plans. The middle layer is edge computing service sub-center that using distributed load balancing. At the bottom of the model is the infrastructure layer of mobile health care terminal devices carrying various sensors. The health signals detected by the sensors are converted into data. All of the data sets are delivered to the edge computing service sub-center according to the specified security policy, and real-time health status guidance will be provided to users. The users can no longer simply view the detected data, but also can see the conclusions based on medical testing standards. This will help users keep track of their physical condition, rather than looking at daily measurements. The cloud computing center is responsible for storing decision data and conducting correlation analysis to provide early-warning services for community hospitals. Results: The framework of cloud computing and edge computing designed for health care, is used to make up for the deficiency of cloud computing. Realize multi-source and heterogeneous data of health care fusion in edge computing. Conclusions: The three-layer mapping model of cloud computing and edge computing fusion mode can provide real-time and reliable services for intelligent health care services, but the privacy protection of users’ health data and service management need further research. Acknowledgements: This word was supported by the Jilin provincial department of education science and technology research project (2017LY505L15) and Jilin province social science fund project (2017M21). Objective: The senior citizen is a special type of consumer group, of which the physiological and psychological functions are gradually degraded. Therefore, the packaging design of the elderly medicine should give priority to ensuring the safety of use. In the meanwhile, the physical features and mental characteristics of the elderly should be considered. Last but not the least, the design should be scientific and reasonable, with more humanism care to those golden-agers. Methods: Based on the visual characteristics of the elderly, through comparing and analyzing the packaging of geriatric drugs and common ones. Then, finding out the key points in the information visualization design that can improve the communication efficiency from. In that way to understand the preferences of the elderly for the visual design characteristics of pills. thus, involved the humanized consideration in the design of medication packaging. Result: Humanized design is a people-oriented design. The elderly medicine packaging design should be based on the elderly, in line with the physiological and psychological characteristics of the elderly. This research firstly analyzes the basic functions that should be possessed by the common drug packaging for the elderly, and then extends the functional design to the use behavior. It concludes that the humanized packaging corresponds to the patient\'s intuitive perception of the medicine packaging, thus making the design more user-friendly; Based on the characteristics of the drug packaging design information transmission function, we can find out the key points that can improve the communication efficiency from the information visualization design. At the same time, this research discusses the green packaging design of the elderly drug packaging, so that the pharmaceutical packaging can protect the nature more. Finally, the specific scheme for realizing the humanized design of pharmaceutical packaging for the elderly is introduced. Conclusion: Through the research on the visual form of medicine packaging for the senior citizens, it can better protect the health and safety of the elderly, standardize the pharmaceutical production enterprises, enhance the practical and convenient functions of packaging, and accelerate the healthy development of the industry. The humanized design of packaging is not only a form of promotion of medicines, but also a reflection of resection and care of the whole society to the elderly. Automatically alarm clothing research on healthy emergency of long-distance interconnection Kaisheng Zhang, Lin Gan, Xiaojuan Dang Shaanxi Fashion Engineering University, Xi\'an, Shaanxi 712046, P.R. China Objectives: With development of computer technology, network communication technology and integrated technology, preparing intelligent clothing with automatic alarm is imperative. In this work, automatic alarm clothing for long-distance interconnection emergency has been studied to supervision based on internet technology and computer technology and utilized clothing as design carrier. Methods: A series of software and hardware should be applied in automatic alarm clothing for long-distance interconnection emergency. The first is algorithm of information processing. In the process of algorithmic writing, both robustness of algorithm and influence of algorithm to complex environmental conditions should be considered. The second is hardware of information processing. Hardware will cause effects on the speed of information processing, and stability and anti-risk property of the whole system. The third is secrecy system. With arrival of big data era, more people have realized the importance of data security. Information leakage has become an inevitable problem. Automatic alarm clothing for long-distance interconnection emergency is a kind of networking equipment. In the process of network connection, personal information may be leaked and this will bring huge economic loss for users, sometimes even irreversible personal loss. Thus making a good secrecy system is very important to protect personal information when using automatic alarm clothing. Results: Information transfer is very necessary after information processing, and stability and reliability of internet should not be ignored during design process. At present, the applied communications protocol family is TCP/IP family, it concludes several different kinds of information transformation modes. Network connection function is the basis for information transformation. Setting server is very necessary for automatic alarm clothing for long-distance interconnection emergency, since it can provide more convenient service for users. Conclusions: Automatic alarm clothing for long-distance interconnection emergency can meet the special requirement of using, and good for future research. During clothing design process, designer can refer three aspects suggested in this work, namely information collecting, information processing and information transfer, to make good design of clothing and hardware. Laboratory Management Center, Shenyang Sport University, Shenyang, Liaoning, China 110102 Purpose: With the approach of the Beijing Winter Olympics in 2022, Chinese people\'s enthusiasm towards ice and snow is heating up year by year, but the number of knee injuries caused by skiing every winter is increasing. This study focuses on the snow plough turn of the mass ski beginners, and conducts finite element analysis on the lateral knee cartilage during the plough skiing process, aiming at revealing the mechanism of knee injuries in mass ski beginners. Method: 200 mass ski beginners (100 males and 100 females) are randomly selected as the research objects, and image analysis is applied to analyze the lateral knee joints of the research objects in the snow plough turn, to get the angle range and gravity center velocity of sliding knees and turning knees; the stress of the position of knee cartilage is analyzed by establishing a human mechanics model, to set the boundary conditions for finite element analysis; CT and MR testing is conducted on a volunteer to obtain DICOM data, and the three-dimensional reconstruction of knee tissue and cartilage tissue is performed and imported into Abaqus finite element software in IGES format to complete the simulation calculation. Results: The kinematic data of 181 volunteers (87 males and 94 females) are obtained, average height H=1.67±0.09m, average weight (including ski equipment) m=68.42±14.64kg, gravity center speed v= 1.67± 0.38m/s, sliding knee angle α=146.38±12.37°, and turning knee angle β=139.63±12.32°. Through the mechanical model analysis, and combined with the sports biomechanical research method, the upper part of the knee tibial plateau accounts for 85.6% of the total. In the case of the turning radius R=0.6m, the coefficient of friction of the snow μ=0.1, and the angle of the sliding slope θ=13°, the lateral knee joint of snow plough turn is subjected to gravity G′=402N and receives centripetal force F′=195N. According to the results of finite element calculation, when the lateral knee joint is 140˚ and the descending center of gravity speed is 3m/s, obvious stress concentration occurs in the medial meniscus and rear feet of lateral knee joint, stress concentration occurs in the middle and posterior part of the lateral meniscus, and stress concentration also occurs in the middle of the lateral margin of the lateral tibial cartilage. Conclusion: There is stress concentration in the medial and lateral meniscus and lateral tibial cartilage of the knee joint; beginners are prone to fatigue due to the lack of special strength of the legs, so they should adopt standard movement and reasonably adjust exercise intensity; the sliding speed should be less than 3m/s, and the turning radius should be increased as much as possible, so as to effectively reduce knee cartilage damage during snow plough skiing. Fund project Key Special Project of National Key Research and Development Program \"Technical Winter Olympics” (2018YFF0300502). Comparative study on the influence of biogas fermentation from food waste among inoculum from different acclimation methods Jin Zeng, Rui Xu*, Yuan Dai, Yage Yuan School of Energy and Environmental Science, Yunnan Normal University, Kunming, China; *Corresponding author Background: According to survey, which shows that China\'s catering revenue in 2017 was 3689 billion RMB, which raised 10.1% compared with the previous year. At the same time, it is imperative that a large amount of food waste (FW) need to be generated. Therefore, this study was based on Yage Yuan\'s previous experimental results, selecting the inoculum from different acclimation methods and with different acclimation time to implement the batch fermentation of biogas production from FW, and trying to explore its impact of inoculum on the potential of biogas yield under mesophilic condition as well as seeking a way to acclimate high quality inoculum. Methods: 3 different acclimation methods (0.5g/d FW, continue to feed for 32days and 680mL/g phosphate; 1g/d FW, continue to feed for 16days and 1360mL/g phosphate; 2g/d FW, continue to feed for 8days and 2720mL/g phosphate) and 4 different acclimation time (8days, 15days, 21days, 44days) were designed, and the anaerobic batch fermentation experiment was conducted with 1 control group (300mL unacclimated inoculum, 16g FW and add water to 850mL) and 12 experimental groups (120mL acclimated inoculum, 16g FW and add water to 850mL). Results: Research results showed that both the daily gas production and cumulative gas production of experimental groups were higher than control group, and it could start quickly and operate stably. Among, the group with 1360mg/L phosphate, 1g/d FW feeding and being acclimated to 15th day was the most feasible and adoptable measurement method for the biogas fermentation from FW. Conclusions: (1) The acclimated inoculum by adding phosphate and FW can ensure faster start-up speed and stable operation of biogas fermentation from FW than unacclimated inoculum, which greatly improves the efficiency of biogas fermentation. (2) The multiple comparison analysis by Duncan method shows that both acclimation method and acclimation time of inoculum can affect potential of biogas fermentation from FW, indicating that different inoculum used for biogas fermentation from FW had a significant impact on the maximum daily gas production and cumulative gas production(P 0.01). (3) In the future, when selecting the inoculum for biogas fermentation from FW, the inoculum in the stable phase of acclimation process and in the middle phase can be selected according to the variation of daily gas production during the acclimation of inoculum, thereby improving the efficiency of whole reaction and shortening reaction time. (4) The experimental group with 1360mg/L phosphate, 1g/d FW feeding and being acclimated to 15th day is the most feasible and adoptable measurement method for the biogas fermentation from FW. Acknowledgments: This work was supported by the National Nature Science Foundation of China (Grants No. 41561108). Research on psychological health development of college students in the new era from the perspective of micro media Jingjing Gao Xi`an Fanyi University, Xi\'an, Shaanxi 710105, P.R. China Objectives: Micro-media has been a carrier and platform for both public opinions transmission and information distribution for students. Colleges and universities must follow the pace of micro-media innovation and enhance the research of the psychological development of college students. Methods: Traditional college mental health education has been realized through two kinds of approaches: the first is offering metal health education courses for college students; the second is professional psychological counseling for college students. Mental health educators in colleges and universities should pay attention to the problems and events that college students participate actively in discussions through micro media platforms, and comprehensively understand mental health status of college students. Meanwhile, timely guidance and intervention should be given to public opinion, and preventive measures should be taken. In the environment of micro media, every person can be producer and consumer of information. Changing one-to-one psychological counseling and one-to-many traditional education mode into one-to-many and many-to-many micro media communication mode, integration of resources and information of mental health education work can be realized. Results: Micro media has become a forward position of metal health education work for college students. Through a variety of information feedback channels, college should hold the psychological dynamics of college students timely, conduct public opinion guidance, and enrich the content and form of psychological health education, thus can improve the pertinence and effectiveness of the psychological health work of the college students. Conclusions: Micro-media is a network platform that integrates information and interactive communication. It opens a new area of metal health education work, and also brings new challenges. Research the psychological health development of college students in the new era under the perspective of micro-media is significant for educate psychological health educators to enhance their media literacy, establish an equal dialogue between teachers and students, cultivate micro-media students management teams, guide public opinion on campus, and create good psychological health educational micro-environment. Acknowledgements: Supported by a project grant from Shaanxi philosophy and social science foundation project (Grand No. 2015N023), and college-level scientific research team project (XFU17KYTDC04). Objectives: In recent years, due to lack of top-level design and system arrangement, the planning and construction makes rapid progress. However, inevitable situation happens, like extensive development and construction, excessive reliance on investment, and malposition of software and hardware. The unordered and unhealthy current situation of smart cities is worth considering. So discussion of healthy developing mode for smart city is imperative. What kind of speed, scale, rhythm and mode can make city become healthy and ordered has become a first problem to be solved. Methods: Globally, two main stages were experienced during construction of smart city. One is scientific smart city and the other is administrative smart city. The key point of the former is advanced research of information and communication technology as well as infrastructure construction. As for the latter, the key point is the application of information and communication technology in government management. The smart city construction of China copied western mode to a great extent, mainly based on geographic information system, three-dimensional simulation and database establishing. However, the reason why unhealthy situation has formed is scientific and administrative smart city mode are all based on western ration culture, and treating smart city construction equally to techniques, capitals and managements. This is not only against the development trend of cultural city and purpose of humanistic city construction, but also against the city essence of creating \"value, significance and dream”. Results: A series of problems exist in the traditional urban management method such as higher administrative cost, extensive administrative approach, single admin principal, outdated information gathering and process mode. Applying solution of smart city will improve sustainable properties of city, and establish a goal system with infrastructure construction, public management, public life environment and city development ability. This will provide basis for subsequent researches. Conclusions: Smart city is different from industrial societal city, which leads to the consequence that the traditional method is no longer applicable. Scientific theoretical direction and practical adjustment should be utilized to promote a healthy development of smart city. Acknowledgements: This work was supported by the Science and Technology Plan Projects of Shaanxi Provincial Department of Education (Grants No. 18JK0933). Objectives: With the development of society, people\'s demand for medical care is also increasing, they desires for more efficient medical services. Due to limited medical resources, a lot of mistakes or misoperation happened every day in the hospital such as rapid identification of patient identity, taking drug infusion bags by mistake. The most common medical accident is a health care worker hangs a bottle on a patient by mistake. Methods: In order to solve the above problems, according to practical business requirements, a solution was proposed to avoid such medical mistakes. Based on RFID technology, the design provided a system for preventing medical workers from accidentally hanging a bottle. The medical workers compared the information of the target patient and target patient\'s bottle through RFID technology and observe if it is accordant or not. If it is accordant, medical workers will continue taking infusion to target patients. Such will avoid hanging wrong bottles to the patients, solved nursing malpractice effectively and reduced mistakes, thus improving working efficiency and nursing quality. Results: The system design of preventing the medical staff from hanging the bottle by mistake, and solve problems like closed loop execution process of medical orders, the safety of patient data and information, and the occurrence of medical accidents. With the development of RFID technology and wireless network technology, laptop, handheld device, etc. has appeared in people\'s view, the information system of patient will become digitized, intelligentized and portable. This RFID-based preventive medical personnel\'s system design is of great practical significance for avoiding medical accidents and building a harmonious society. Conclusions: In recent years, with the development of hospital informatization, many hospitals are considering how to extend medical services to patients’ bedside. The RFID-based preventive medical personnel\'s system design will change the working model of the original medical order system, embody the service concept of patient-centered, and improve the work efficiency and management level of medical staff, and finally realize the situation of preventing medical mistakes. 1School of International Tourism and Business Administration, Lijiang Teachers College, Lijiang, China, 674199; 2Lijiang Tourism Research Center, Lijiang Teachers College, Lijiang, China, 674199; 3Heritage Tourism Research Base, Yunnan Provincial Tourism Development Commission, Lijiang, China, 674199 Objectives: In recent years, health tourism has been becoming more and more important in tourism development in the new normal of China\'s economy. This research has been carried on to sort out and study the literature of China\'s health tourism since 1998 in order to satisfy the needs of studying the situations further and offering some scientific suggestions and guidance for the following studies on health tourism. Methods: Taking health tourism as the theme, 441 research papers, published in China\'s journals, were taken as the samples and were analyzed metrically according to bibliometric and textual analysis on the basis of the data from CNKI. Results: 1. Study on China\'s health tourism is at the beginning in the research process. Only 69 papers were published in this field from 1998 to 2014 and 372 were published from 2015 to now. Furthermore, the research process could be divided into two phases: emerging and exploring phases. 2. The authors of the literature are mainly from CTM universities and comprehensive universities. 3. Only 0.23% study on health tourism are funded by research funds, which reveals the situations of low rate of funding, the shortage of capital and investment in this field. 4. Only 24.49% of the literature was cited, showing there exist low rate of citation per paper and shortage of high-quality papers. 5. Researches on this field mainly focus on health tourism, resources of health tourism, tourism of health-maintenance, tourism of health care etc. 6. Only 5.89% of the literature was published in the key journals including SCI, EI, CSSCI and CSCD, indicating that high-level research on health tourism are in deficiency. Conclusions: In consideration of insufficiency of research on China\'s health tourism, effective measures are taken to meet the needs of its promotion: 1. To strengthen comprehensive theoretical research; 2. To enhance multi-perspective interdisciplinary research; 3. Promote the study on planning and policies of health tourism; 4. To consolidate the construction of research group; 5. To improve the investment to the study on health tourism. Acknowledgements: This work was supported by the National Natural Science Foundation of China (Grants No. 41361037&41671147) and authorized fund of Yunnan Provincial Tourism Development Commission (Grants No. YLHT2018001). 1School of Humanity and Law, Harbin University, Harbin, China; *Corresponding author: Tel.: 15776761752, E-mail address: 271115447@qq.com Objectives: Mass sports is an important part of China\'s socialist modernization, the development of mass sports is an important measure to strengthen the national physique, improve the quality of life and build a harmonious society. Harbin, located in the northernmost part of China, is a metropolis with the highest latitude and lowest temperature in China. Studying the current situation of mass sports in Harbin can not only provide decision basis for local governments to carry out sports work and promote sports industrialization, but also provide reference for the development of mass sports work in northern cities, especially in winter. Methods: This study mainly used the methods of documentation, observation, interview and questionnaire to investigate the exercisers of 15 fitness clubs and gyms in urban areas of Harbin, as well as public places such as park streets. Results: 1. Among the participants of mass physical exercise in Harbin, the age is out of balance, and the middle-aged group is not enough. Because of heavy work load, social intercourse, heavy housework, irregular work schedule and subjective lack of understanding of physical exercise, the participation motivation is not strong, the possibility of breaking physical exercise is greater. 2. The mass sports consumption concept is not strong, at the same time the scientific fitness idea is backward, causes the club, the paid service auxiliary station to develop slowly, the sports industrialization has a long way to go. 3. Mass sports is less organized, and the project selection is influenced by such factors as gender, age and regional cultural characteristics. In addition, the shortage of sports field also restricted the development of mass physical exercise. Conclusions: 1. The society, the government, and the units should change their views, takes the mass sports seriously, especially the middle-aged people\'s sports fitness, and we will give full play to the role of social bodies in guiding and supervising the physical fitness of middle-aged people. 2. Strengthening the publicity and promotion of the national fitness plan by the government, especially the functional departments of sports at all levels, make people gradually establish scientific concept of fitness, change people\'s sports consumption concept, let the health investment is deeply rooted in the hearts of the people, so as to promote sports toward marketization and industrialization. 3. Intensify that organizational strength of sport project, fully exert the advantage of the ice culture in this city, develop physical exercise program with local characteristic, the form of physical organization should adopt welfare type, public welfare type, profit type and so on, in order to meet the needs of different groups. Experimental study on the effect of gymnastics on vestibular system function of college students Qiaoyan Yu1, Chenliang Deng2, Zhixiang Zhang3 1,3Dept. of Physical Education and Training, Chengdu Sports University, Chengdu, China; 2Dept. of Sports, University of Electronic Science and Technology of China, Chengdu, China Objectives: To explore the gymnastics movement to improve the function of vestibular system and the mechanism of resistance to stun, find out the necessity and irreplaceability of gymnastics practices, for young students and the masses to carry out the gymnastics provide certain theoretical support. Methods: The experiment evaluation methods mainly include rotatory test and positional nystagmus test. We divided 50 PE students aged 19 to 20 from Chengdu Sport University into two groups, and each group included 13 female students and 12 male students. According to the syllabus of school gymnastics teaching, some exercises for men include the vault, horizontal bar, parallel bar and floor exercise. The women\'s section includes floor exercise, horizontal bar, balance beam, vault and parallel bar. The experiment was according to the basic conditions of sports professional students, and carry out systematic gymnastics teaching and training to them, the experimental course is designed and operated by myself. Finally, the data measured before and after the experiment were statistically compared and analyzed. Results: Vestibular system is closely related to spatial perception and movement, and it is an indispensable link in human sensory integration system. In the experiment, different movements in gymnastics were selected, such as those with different plane rotation, handstand, swing, space and ground, etc. After an experiment lasting 16weeks (3 times per week), the rotatory and positional nystagmus test was used to test the effects of the experiment, which is to make the client to sit on the swivel chair, head fixed in forward 30°, made the semicircular canal in horizontal position, at a speed of 2seconds a lap to rotate to the right direction (clockwise) to 10 times and stop suddenly, ask client to two eyes forward, with is the swivel chair forward line, observation of nystagmus, dizziness and a straight line, record the data. The pretest results of the experimental group showed that 39 cases induced ocular shock, 42 cases could not walk in a straight line, and 31 cases were accompanied by positional vertigo. After the experiment, it was found that 11 cases induced ocular shock, 15 cases could not walk in a straight line, and 7 cases were accompanied by positional vertigo. These data reflect significant differences before and after the experiment. Conclusions: There are different movements of rotation, jumping, somersaulting and balance in gymnastics, which can change the position of the head and body. Therefore, the long - term gymnastic exercise of certain intensity is important to the stability and function enhancement of human vestibular system. Acknowledgements: This work was supported by the Education Department Scientific Research Project of Sichuan Province (Grants No. 18SA0032). School of Computer and Information, Shanghai Polytechnic University, Shanghai, China Introduction: With the advent of the Internet wave, various e-commerce legend was born, has prompted a large number of relevant industry entrepreneurs into the Internet. And it has produced a large number of development of relevant industries. From basic web design, pictures, beautify the purpose of other classes such as creative design to network marketing, Weibo marketing such as marketing promotion, and program development of website development, APP development etc. So the question comes, you where to find design talents, marketing talent, develop talent. Yourself for a related staff cost is too high, look for the national well-known team service charge is too expensive, the list is too small, big companies look. At this time, it needs a information platform to butt small single the house from top to bottom, Witmart arises at the historic moment. Methods: A. Witmart from the perspective of customers, providing customers with Witmart help, new beginner, etc; B. Witmart name \"pig eight quit” this is one of the four great classics \"journey to the west” is one of the main characters, basic domestic everyone read the classic TV series \"journey to the west”, the name kind humor, easy to remember and fun; C. Special operating mode, after login authentication, can be a reward, tender task, according to the demand that, very human nature; D. Original model, under the mode of \"reward”, \"the tender”, added \"speed dating”, \"mediation” mode. To help users better and more comfortable experience. Results: Witmart crowdsourcing is the largest domestic service platform, has a perfect service system, being the service provider, with 13 million page views per day, Alexa global comprehensive ranking of 4619. In the Internet age, a growing number of Internet companies and the birth of the Internet around the company, the traditional enterprise also began to build their own e-commerce department, it will have a large number of outsourcing service list needs to be done, but have never Witmart if low level service provider, to continue as an industry leader, will be under the wave of the Internet to become a great winner. Conclusions: Witmart is a dedicated service crowdsourcing plat form, this is the place where it is different from other e-commerce sites, and it can be in today\'s era of e-commerce sites flooded the meteoric rise of important rea sons. Web site itself has no real tangible goods, also did not give stores to provide physical sales platform. It focus on the service, the website provide a have some aspect knowledge, technology, talent and need related services companies, organizations, individuals and docking platform. Here, the knowledge is money, technology is money. Pay attention to the current intellectual emphasis on talent idea coincides with mine. China\'s first service crowdsourcing platform of e-commerce sites. Acknowledgements: This work is supported by the Key Disciplines of Computer Science and Technology of Shanghai Polytechnic University under Grant No. XXKZD1604. 1School of Reliability and Systems Engineering, Beihang University, Beijing, P.R. China; 2Department of Aeronautic Electronics, No. 1 Aviation College of Air Force Engineering University, Xi\'an, P.R. China; 3Key Laboratory of Science & Technology on Reliability & Environmental Engineering, Beihang University, Beijing, P.R. China Objectives: Classifying and coding network faults are fundamental and essential for fault diagnosis and prediction. It is difficult to deeply classify and code network faults because of their multiple influencing factors and complex dynamic coupling relationships. So network faults are still traditionally classified according to independent fault causes, such as hardware, software, setting, and etc. in reports of RAND, NRSC, ATIS and some other famous researches. Inspired by ICD (International Classification of Disease), a structural data model containing five classification axes is proposed. Finally, a communication network fault classification and coding case is realized to prove the feasibility and advantages. Methods: Since the multi-axis taxonomy of ICD has realized classification for complex dynamic coupling diseases and there are plenty of similarities in network fault classification, we correspondingly transform disease classification criteria of lesion, symptom, etiology and pathology into fault classification criteria of location, phenomenons, causes and mechanism and further borrow the double rules to describe dynamic coupling relationships of faults. And we refine the detailed contents according to similar things of diseases. Such as, according to etiology, we divide network fault causes into the biological, inborn and etc. and give some examples as computer network virus, low processing ability and etc. Then we propose a structural data model including five classification axes as follows: In communication network, an example fault F3 is described by the structural data set {N20; Pn1; Cb1; Cc1; M2; Da1; Db1} under the following situation: when a network switch (N20) with low processing ability (Cb1) meets big data flow (Cd1) and the configuration (Cc1) is to drop the packages out of time, then the phenomenon of package loss (Pn1) will be obvious. And further if the switch is on the only path from terminal A to terminal B because of the topology design (Db1) and then another fault (Da1) happens due to the resource competition (M2). In the set of {N20; Pn1; Cb1; Cc1; M2; Da1; Db1}, there are more detailed and important information, such as, multiple influencing factors (Cb1, Cd1, Cc1), dynamic coupling relationships (M2) and etc. than the traditional network classification method, in which there is only location information (N20). Results: The example fault F3 has proved the classification and coding feasibility of the model. And more network faults can be detailed classified and coded in the same way. Conclusions: In this study, a classification and coding model based on ICD is proposed to deeply classify and code network faults with multiple influencing factors and complex dynamic coupling relationships. This model could give detailed information of influencing factors and coupling relationships and improve the effects of network fault analysis, diagnosis, prediction and etc. Acknowledgments: This work was supported by the Beijing Natural Science Foundation (Grant No. L171002) and the National Natural Science Foundation of China (Grant No. 61872018 and Grant 61773044). Study on treatment of low permeability oilfield wastewater by microfiltration iridium membrane Chunwei Shi1,2, Tianjun Zhan1, Xing Yu1, Haiyang Wang2, Qingquan Wei2, Liang Qiao2, Honglei Shi2, Yili Duo1, Yanhong Zhang1 1Liaoning Shihua University, Fushun, China; 2Administration Center of the Yellow River Delta Sustainable Development Institute of Shandong Province, Dongying, China Objectives: According to the characteristics of the produced water in the low permeability oilfield, this paper designs an integrated filter skid device for wastewater treatment. This device applies microfiltration membrane technology, as well as the auxiliary process of flocculation sedimentation. Methods: After using this device, the indexes including the content of suspended solids, the diameter of the suspended median particle and the content of oil could met the requirements of sewage reinjection in that oilfield, contributing to alleviate the water resources waste and environmental pollution. Then based on results of the microfiltration process, this paper set up experiments to figure out the factors which influence this process. Results: Focusing on the effects of operating temperature and pressure in analysis, this paper presents the optimum operating temperature and pressure of the metal membrane filter. Finally this paper gives the conclusion of the whole research which is instructive to the problem of the produced water treatment in low permeability oilfield. Conclusions: The metal membrane filter treatment of oilfield wastewater, novel design, good effect, solve the poor treatment effect of oilfield produced water, suspended solids and other indicators to achieve the water quality requirements, and has good application prospects in the field. The metal iridium film a key role on the removal of emulsified oil and oil soluble and suspended solids to micron level, compared with the traditional metal membrane filtration process, the membrane integrated wastewater treatment skid device adopts coagulation sedimentation separation of multi mechanical filtration membrane pretreatment process and tangent spiral water technology, to ensure the filtering effect of metal iridium film. The settlement process was improved at the suggestion in the device on the flocculation, fungicide, increase the flocculation and sterilization process, and then increase the dosing process in wastewater treatment process, adding to the water soluble polymer scale inhibitor treated wastewater. Acknowledgements: This work was supported by the Research Fund for the Foundation of Liaoning Education Committee (L2017LQN002) and the Talent Scientific Research Fund of LSHU (2016XJJ-064). Background: As a practical management concept and method which aims at improvement and evaluation for government performance, performance management has attracted increasing attention by the worldwide governments and being wildly adopted in different countries, and also has gradually employed in our government management. The evaluation for government invested project performance is one of the key indicators for government performance. There is a pressing need to introduce new performance evaluation techniques and methods, in order to overcome the low efficiency of financial capital expenditures basically, to improve the performance level of funds, and to improve the administrative capability of the government. It is also meaningful for facing the emerging financial context and for enhances the benefits of fiscal expenditure. Methods: Due to the variety and complexity of evaluation model for government invested project, a novel evaluation model construction method based on improved grey correlative degree analysis has been proposed. Firstly, the correlation between various indexes of government invested project and the evaluation system has been calculated. Dealing with expert scores based on the Delphi method and assigning the results to the importance score of each evaluation index. Secondly, the correlative degree of evaluation index has been combined with the importance score and the initial indicators have been screened after sorting. Finally, an effective model has been drawn out through analyzing the correlations of primary indexes. This method has been well integrated with the two aspects of knowledge. which are expert\'s evaluation and index correlation. This method also has been well integrated with the expert\'s viewpoint and reflected the combination of qualitative and quantitative analysis. Results: To overcome the problems existing in the performance evaluation system of government invested projects, this dissertation has studied the theory and technology of government invested project performance evaluation index system and decision analysis. Combining quantitative methods with qualitative methods, a scientific performance evaluation model of government invested projects has been established, which is also based on the related technology of artificial intelligence and management science and engineering. This performance evaluation system provides a solid theoretical foundation for the implementation of government invested projects performance evaluation. Conclusions: The system includes a method for constructing an evaluation model based on the improved grey relation grade, an integrated weighting method based on the ideal solution and the combination of the sum of squares, and a method for evaluating the performance of government invested projects based on DEDM method. Acknowledgement: This research was financially supported by Shaanxi Social Science Foundation Project in 2018 \"Research on the innovation ecosystem of new industries in Shaanxi under the background of ‘one belt and one road’” (Grants NO. 2018S30); Shaanxi Provincial Department of education special scientific research project in 2018 \"Research on innovation synergy of Shaanxi emerging industry ecosystem under the perspective of \"one belt and one road” (Grants NO. 18JK0261). Gang Zhang1, Shimin Kang1, Xiangxuan Huang1, Jing Hai2, Mingzhong Ren2, Zhongwu Cai1, Yiyan Chen1, Zhenxi Yuan1 1School of Chemical Engineering and Energy Technology, Dongguan University of Technology, Dongguan, China; 2South China Institute of Environmental Sciences, Ministry of Environmental Protection, Guangzhou, China Objectives: Polychlorinated dibenzo-p-dioxin and dibenzofuran (PCDD/Fs) emissions from simple waste incinerators (SWI) may cause health risks in nearby people and are thus subject to stringent regulations. The aim of the study was to evaluate environmental risk of PCDD/Fs emission situation of the SWI under uncontrollable combustion condition and nonexistent air pollution control device (APCD). Methods: The investigation was carried out in a simple small-scale waste incinerator in the suburban area of South China. The SWI, with a combustion capacity of 400kg/hr, operated for 8h per day. The SWI was equipped with a grate and an air inlet at the bottom of furnace. Waste was fed from the top of the furnace. Combustion air and flue gas for the incinerator are highly dependent on natural draft. The incinerator was not equipped with APCDs and gas cooling systems. To evaluate the formation, emission and distribution of PCDD/Fs from different components constituting the dioxin inputs and outputs in the SWI, the flue gas, stack fly ash, bottom ash and solid waste samples were collected thrice from the corresponding sampling points. To assess the influence of PCDD/F from the SWI on the plant environment, the soil and waste water by washing ash of the plant area were also sampled. Results: The stack gas and fly ash presented average PCDD/F levels and emission factors of 3.6ng I-TEQ/Nm3 and 189.31μg I-TEQ/t, 6.89ng I-TEQ/g and 137.85μg I-TEQ/t, respectively, much higher than those from large municipal solid waste incinerators (MSWI). The difference of the congener profiles between input side and output side indicated PCDD/Fs from the feeding waste weren\'t thermally decomposed in the furnace, and were generated mainly by heterogeneous catalytic reactions in the low temperature post-combustion zone. A maximum production of PCDD/Fs about 324.66μg I-TEQ were achieved per one ton weight of solid waste incinerated under the worst and uncontrollable combustion conditions, which may be attributed to waste component, uncontrollable combustion conditions, and inexistent APCDs. The mean PCDD/F concentrations for soil and wastewater by washing ash was 318.33ng/kg and 151.04ng/L, respectively, and the corresponding mean I-TEQ concentrations were 24.08ng I-TEQ/kg and 10.52ng I-TEQ/L. The results obtained do indicate that the impact of PCDD/F emissions from the SWI on the plant environment was quite significant. These results can provide useful information for making clear about the real situation of dioxin generation from the uncontrollable combustion of solid waste and developing better control strategies for the SWI. Acknowledgements: This work was supported by the National Natural Science Foundation of China (21606045), Natural Science Foundation of Guangdong Province of China (2017A030313084, 2017A030313261), and Young Innovative Talents Program from Department of Education of Guangdong Province (2015KQNCX163). 333 Long Teng Road, Sino-Korea Multimedia School, A117 SUES, Shanghai, China 201620 Backgrounds: Information design has become more and more popular these days as we have step into a smart phone new era. The mobile phone is changing the mind set of human beings since we are only willing to spend little time on each single page. And this requires all the information to be very direct and visually clear. Addition to it, Information design has been widely used in public service system, not like in Western world, China is developing rapidly only in these few decades, all the fundamental instruction has been well constructed. However, there are many more can be improved in information design. Methods: We have conducted several field trip to collect necessary information. Taking Metro system in Shanghai as an example. We select one major conjunction metro station Xu Jia Hui, and one bus, airport, metro station Hongqiao. During the trip, we have interviewed many passengers, staff members working in the station. We also took photos, recorded our survey result Results: The general sign and signal design is clear, however, the information design in Chinese public system lacks something more original. There are no unique visual symbol or visual taste in each single city. And some of the design elements are not consistent Conclusions: As we can see from the survey result above, it requires more unite and more unique design to be used in Chinese Pubic service information design system. Analysis of the double cross sliders cycloidal-pin wheel reducer based on finite element method Lei Song1,2, Shunke Liang3, Zhenxing Zheng1, Feixin Chen1 1School of Mechatronic Engineering, Guangdong Polytechnic Normal University, Guangzhou, China; 2South China Robotics Innovation Research Institute, Foshan, China; 3Guangzhou College of South China University of Technology, Guangzhou, China Objectives: Double cross sliders cycloidal-pin wheel reducer, because of its high-efficiency, high-ratio torque delivery, and compact volume, has become one of types precise transmission mechanism in medical equipment. The analysis of stress and deformation calculation made clear the influence of key components on the whole reducer, providing the technical reference for the double cross sliders cycloidal-pin wheel reducer in design, manufacture and application. Methods: This paper use FEM to fulfill the research object. The 3d model of the reducer established in Pro/E was imported into ANSYS. The FEM model was divided freely using tetrahedral elements, boundary condition and material parameters are set up, and calculation results are solved out. Results: According to calculation results, the maximum contact stress of the cross sliders mechanism is 377.9MPa, and its maximum deformation is 0.0078mm. The maximum contact stress of the cycloid-pin wheel mechanism is 32.236MPa, and its maximum deformation is 0.003367mm. The maximum contact stress between the eccentric shaft and the eccentric bearing is 105.11MPa and only half of the rollors have contact stress. The maximum deformation of the eccentric shaft mechanism is 0.09027mm, which appears at the eccentric shaft section. Conclusions: Under the rated load, the maximum contact stress and equivalent stress appear on the cross slider mechanism. However, the maximum contact stress and equivalent stress are both less than the strength of the corresponding materials, which indicates that the strength is enough. The stiffness of the eccentric shaft is smaller than the cross slider and cycloidal-pin wheel under the rated load. The maximum deformation is 0.090mm, which will have great influence in the reducer transmission precision. Therefore, the materials of eccentric shaft and manufacturing processes should be reasonable selected to improve its stiffness. As for cycloid wheel, in order to improve the rigidity so as to improve the transmission accuracy, it needs to choose the modification methods and modification reasonably according to the deformation. Acknowledgements: This work was financially supported by the Guangdong Natural Science Foundation (Grants No. 2018A030313509) and Major Special Project of Guangdong Provincial Science and Technology Plan (Grants No. 2017B090910007). A study on factors affecting online game addiction—Using the quantitative meta-synthesis method Pengchun Lin1, Hsinke Lu1, Shaoyu Lu1, Kuochung Chu2 1Chinese Culture University, Taipei, Taiwan; 2National Taipei University of Nursing and Health Sciences, Taipei, Taiwan Objectives: In light of this development, plenty of scholars had conducted researches on using behaviors of online game; the related results showed that online games would give players satisfaction, but they would derive problems related to online game addiction. In this study, the phenomenon of online game users’ internet addiction of different gender was investigated. Methods: In this study, data was analyzed and calculated by tools of Meta-Synthesis and related formulas, and it was further analyzed and interpreted for publication. The researcher utilized the research data published in Taiwan and China during this decade as the basis and used the quantitative meta-synthesis to sift 14 studies utilizing internet addiction instrument to investigate the influence of users’ gender on their tolerance, compulsion, withdrawal reaction, time management, interpersonal relationship, and health problem of online game addiction. Results: Gender of those users had significant influences on each dimension. Researchers expected that the results could provide references for researchers and education institutions related to this field so that gender differences of online game users’ tolerance, compulsion, withdrawal reaction, time management, interpersonal relationship and health problem of online game addiction could be understood. In this study, those values were integrated into the CMA analysis by their dimensions; this meant that \"tolerance” dimension among CIAS influenced males more, and the average effect \"Correlation” column Zr=+0.100(p 0.001) reached the significant level. For withdrawal reaction, \"withdrawal reaction” dimension among CIAS influenced males more, and the average effect \"Correlation” column Zr=+0.109(p 0.001) reached the significant level. For compulsion, \"compulsion” dimension among CIAS influenced males more, and the average effect \"Correlation” column Zr=+0.112(p 0.001) reached the significant level. For healthy interpersonal relationship, \"healthy interpersonal relationship” dimension among CIAS influenced males more, and the average effect \"Correlation” column Zr=+0.110(p 0.001) reached the significant level. It showed that \"time management” dimension among CIAS influenced males more, and the average effect \"Correlation” column Zr=+0.150(p 0.001) reached the significant level. From the above data, it could prove that males were influenced more than females were in online game addiction. Conclusions: From the data and results of these studies, it was known that dimensions of online game addiction were all different. In a single study, values (Hedge\'s, F, t or ) could be compared; as the value of a certain dimension was higher, it meant that it had more influence on online game addiction. The results showed that the standard deviation of mail higher than female was as following, tolerance =0.20, withdraw reaction=0.22, compulsion=0.23, healthy interpersonal relationship=0.22 and time management=0.30. From the above data, it could prove that males were influenced more than females were in online game addiction. Evaluating communication costs in mobile automotive networks adopting a tree-based model HuaYi Lin Department of Information Management, China University of Technology, Taiwan; E-mail: calvan.linmsa@gmail.com Objectives: This study proposes an evaluation methodology to estimate the communication cost on a node-based V2V (Vehicle-to-Vehicle) and a tree-based V2V in automotive networks. This investigation deduces general models to demonstrate that the tree-based approach outperforms the normal node-based approach in terms of communication costs. Additionally, the proposed model is well suited for implementation on numerous vehicles in large automotive networks. Methods: In automotive networks, the node-based V2V is a plain model for connecting each vehicle in the absence of infrastructure. However, the communication cost quickly becomes excessive when numerous vehicles are involved. This study would like to propose a tree-based communication model to reduce communication costs. All the vehicles are grouped into an x degree tree, where x depends on the number of vehicles, the greater the number of vehicles, the larger the value of x. Finally, this study compares the communication cost of the tree-based with the node-based infrastructure. This study assumes a topology that the automotive network has n moving vehicles which are located on an x degree tree. Here we divide x vehicles into each group, and represent the entire vehicles as an x degree based tree, as shown in Fig.1. The parent vehicle is responsible for the group communication with the children vehicles. Additionally, the root node takes charge the communication of the entire vehicles. Results: If the vehicle V1 (located in group A) would like to communicate with the vehicle M, the original communication cost is 7 hop counts from vehicle V1 to VM. In the proposed scheme, we group x (=5) vehicles as a group, meanwhile we choose a vehicle as the parent node to take charge the group communication. Thus the communication path is V1→HA→PA→HB→VM, and the communication cost equals 4 hops. Even if V1 would like to communicate with VN, the communication path is V1→HA→PA→RA→PB→HC→VN, and the communication cost is only 6 hops not N-1 hops. When we increase the degree of the tree, this study infers the higher the degree, the lower the communication cost. This study computes the height of the tree that equals logxN, where x is the degree of tree and N is the number of the vehicles. This study deduces that the communication cost value is 2≤ communication cost ≤ 2logxN. When the number of vehicles becomes vast, our mechanism can efficiently reduce the communication cost. Conclusions: The tree-based communication model reduces the communication cost. We exploit an x degrees tree to represent the communication model of the entire vehicle infrastructure. When a vehicle would like to connect the others, this study uses the parent vehicle and the root vehicle efficiently connect the destination vehicle via the proposed model. Moreover, this study infers that the communication cost is 2≤ communication cost ≤ 2logxN. From the deduction, our proposed model indeed reduce the communication cost for the vehicle communication. Land use pattern based on SLUTH scenario simulation: balancing the lake ecological environment and urban sprawl Jie Jin1,2, Hong Zhang1, Yue Zhou2 1Collaborative Innovation Center for Development and Opening up of Southwestern Frontier and Mountainous Areas, Yunnan University of Finance and Economics, Kunming, China; 2Faculty of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming, China Objectives: Because urban sprawl in the plateau lakeside has a serious impact on the lake ecological environment. The key to balance or coordination is to find an intensive ecological land use pattern between the plateau lake ecological environment and regional urbanization. It is imperative to Explore new ways of lake ecological protection in plateau Lake and urbanization area from the perspective of land use. Methods: The SLEUTH model proposed by Clarke (1998) as an implementation of Cellular Automata (CA), which is a mature method to predict land use change. This research designs, simulates and evaluates four land use patterns in Dianchi watershed, which considers control indexes such as ecological environment and intensive in the model to adjust the speed and direction of development in different regions, specifically, (a) Landscape ecological constraints. (b) Intensive land use. (c) Develop the center town. (d) Optimize rural residential places. (e) Compactly grouping development. The research implemented and examined the four scenarios of the dynamic process of urban sprawl in different periods in Dianchi watershed. Results: First, development scenarios ecologically-constrained development (II) and ecologically-constrained intensive land use development (IV) under ecological constraints save more land areas for development as much as scenarios development without constraints (I) and intensive land use development (III) without ecological constraints, this shows that ecological constraints are conducive to the conservation and intensive use of construction land, and thus the urban sprawl (especially the City-Town-Village or CTV)expansion under the scenarios IV is the best land use pattern for Dianchi watershed. Secondly, development scenario III becomes larger than I as the cumulative percentage declines, this shows that urban global urbanization promotes intensive land use by the land use policies of optimization and adjustment. However, the CTV expansion without ecological constraints has a probability of 70% to lose control and encroach on ecological land of plateau lakeside, i.e.. Finally, the ecosystem of Dianchi will deteriorate further if the existing urban sprawl without any constraints (i.e., development scenario I) in Kunming. Conclusions: A well-balanced land use pattern based on SLUTH scenario simulation in the Dianchi watershed are summarized as follows: First, land use optimization strategy through ascending ecosystem services or improving ecological economic carrying capacity, controlling the urbanization scale of Dianchi watershed appropriately, adjusting the land use structure of urbanization, optimizing urban function and industrial structure, implementing the zoning control of ecological construction and environmental protection measures, and the urban planning should delineate ecological red lines. Secondly, Strengthening the management mechanism of land intensive use. Scientifically and rationally determine the development model of CTV, control the speed and scale of construction land expansion, strengthen the construction of Dianchi watershed ecological protection, industrial upgrading and transformation and so on. Objectives: It is difficult to match accurately for similar images. Because there are almost too many similar local feature points on images. According to SIFT (Scale-invariant feature transform) algorithm, the image can\'t be distinguished effectively. Although SIFT is the one of the most famous algorithms in image matching for its robustness of scale changes, view changes, light changes and rotation changes. The probability of extracting the same local feature points on each of similar images is too large to distinguish them. It is necessary to get different local feature of similar images accurately, this means that it is used to distinguish similar images effectively. Methods: Applying Co-Kriging model to optimize local feature points and extract the difference of similar images accurately. First, computing the value of each point of each layer image in the difference of Gaussian pyramid. Then constructing a Co-Kriging model. Finally, optimizing the local feature points and enhancing the accuracy of key points so that extracting the difference of similar images. Results: Because this method that applying Co-Kriging model to optimize and extract local feature points is different from other methods. We use Matlab to simulate and transform the data of network images. Applying Co-Kriging model to optimize local feature points and extract the difference of similar images. Using the enhanced method to distinguish similar images, the success rate is 71.3% by matching 300 times. But success rate of original algorithm is only 55.8%. The result of experiments show that the proposed method is more effectively than the original SIFT algorithm in distinguish similar images. Conclusions: The local features of the image is used to distinguish different images. In this paper, an enhanced SIFT algorithm based on the Co-Kriging model is proposed. According to the proposed method, generating the initial key of different scale images, then constructing a Co-Kriging model to optimize the local feature points and enhancing the accuracy of key points so that extracting the difference of similar images. The features of the image extracted by the proposed method is more stable than that of the original algorithm. The advantage of the proposed method is used to distinguish similar images effectively. But it takes more time in the optimizing step. Acknowledgements: This work is supported by National Natural Science Foundation of China (No. 51775472). 1Computer of Science and Engineering. Department of Computer, Changchun University of Technology, Changchun, China; 2National Engineering Laboratory, CRRC Changchun Railway Vehicles Co., Ltd., Changchun, China; 3Department of Automation, Tsinghua University, Beijing, China Objectives: This paper constructs a communication network routing decision method based on evaluation model, and uses dynamic autonomous data transmission optimization scheme under restricted conditions to ensure network connectivity, enhance link robustness, reduce network energy consumption, and prolong network life. In the period, a more comprehensive and complete data transmission research system for communication network transmission systems will be established to improve the efficiency of data links involved in wireless communication networks, reduce information transmission delays, and establish a final implementation to construct complex and variable routings suitable for communication networks. The goal of the data forwarding mechanism of the transmission system in a multi-restricted environment. Methods: In the wireless communication network, a Markov decision ant colony data forwarding link selection scheme based on multi-parameter equalization is established, which considers the link quality, node residual energy and node neighbor number in the network, and is different according to the actual application scenario. Weight, using Markov routing decision model to evaluate the nodes in the communication range of the node, and use the evaluation result as the basis for the ant colony to find the optimal path, thereby improving the overall performance of the network, extending the network life cycle, and achieving the optimal selection of the global routing strategy. Results: Compared with the traditional communication network node transmission strategy, the packet delivery rate of the node is increased by about 15%, the information transmission delay of the node and the node is reduced by about 10%, the energy loss of the network is reduced by about 20%, and the network life cycle period is used. Increase by about 15%. Conclusions: A more efficient and less lossy communication network routing decision method. Data forwarding strategies are feasible in complex environments and should be considered. Acknowledgements: This work was supported by the Science and Technology Department Foundation of Jilin Province (Grants No. 20160520020H). This work was supported by the Education Department Twelfth Five-year Science and Technology Research Foundation of Jilin Province (Ji Kekehezi 2016 No. 342). Research on construction and scheduling optimization model of intelligent industrial energy management system Yi Liu, Jian-hui Jian* School of Economics and Management, NCEPU, Beijing, China; *Corresponding Author. Introduction: With the continuous improvement of the theory and practice of smart grid and load side demand response, intelligent power utilization has also attracted more and more scholars’ attention. Compared with household and commercial users, industrial users consume large amounts of electricity, have a high level of informatization, and are more complicated in their use of electricity. This provides a broad platform for the development of industrial intelligent power utilization. This paper firstly constructs the industrial intelligent power management information system architecture based on demand response, expounding the main function of four function modules which are power dispatching module of production equipment, power control module of temperature control equipment intelligent, photovoltaic power generation management module and decision module of purchase and sale. Then, combined with the electricity usage behavior and characteristics of industrial users, this paper proposes an intelligent power dispatching optimization model with peak load and user cost as the optimization objective. Finally, the paper has verified the validity of the model. Methods: This paper constructs four modules of industrial user intelligent power management system based on the principle of demand response. On this basis, an optimization model for intelligent power dispatching of production equipment was established. The model uses the set theory to describe and define the power task, and establishes a multi-objective optimization model with the load peak and cost of electricity as the optimization objectives, and proposes a solution algorithm based on hybrid coding genetic algorithms. In addition, different algorithm coding and algorithm operation strategies are designed for different production equipment electricity consumption tasks respectively, so as to solve the single objective and multi-objective optimization model for the intelligent electricity consumption task scheduling optimization of industrial production equipment. Conclusion: Based on the four functional modules of industrial user intelligent power management system, this paper proposes a production equipment scheduling optimization model. The research results show that the application of the optimization model can not only help industrial users to achieve the goal of demand response, but also reduce the electricity cost. This brings good technical and economic benefits, and greatly enhances the scheduling optimization function of intelligent industrial energy management system. Acknowledgement: This paper is supported by supported by \"the Fundamental Research Funds for the Central Universities (2018MS062)”. 1Changchun University of Science and Technology, Changchun, Jilin, China; 2Northeast Normal University, Changchun, Jilin, China Objectives: Movement human actions recognition has become a research hotspot in computer vision, and has been widely used in many fields. We mainly investigate and study the key techniques involved in the process of human motion recognition. The aim is to achieve fast and accurate detection for the case of small samples or short video sequences, or robust detection in complex environment. Methods: We have surveyed the academic results in the last twenty years on foreground detection, and none of the models fitted all complex environments. We fused the background subtraction with improved three-frame difference to realize online detection and segmentation for the region of interest. The methods of feature expression and extraction mainly included body models, template feature, global descriptor and local descriptor, which had their own advantages and disadvantages. On the basis of the moving object region detection, we proposed the expression of optical flow energy image, and used a deep convolutional neural network (DCNN) for feature extraction. The main classification and recognition algorithms included template-based models, generative models, discriminative models, and so on. At present stage, we designed the SVM classifier and performed experimental analysis on public data sets. The experiments used cross validation. We split data sets into training and validation data. Pick 30% from each set for training and the remainder 70%, for the validation data. Randomize the split to avoid biasing the results. Results: The fusion of background subtraction and three-frame difference ensured detection accuracy and realized real-time detection for moving objects region. The optical flow energy image based on region reduced the computation and improved the time efficiency. The DCNN feature contained rich information, which had a good ability to distinguish. The accuracy of the experiment on Weizmann was 98.3%. The accuracy on KTH by the same solution was 92.7%. The main factors that affected the recognition accuracy were the effects when the lens was pulled closer, far away or when the camera was moved slightly. On CASIA, we split video data into single-person behavior and complex interaction anomaly. The classification accuracy of single-person behavior was 98.7%, complex interactive anomaly recognition was 93.2%, and the accuracy of mixed data was 95.6%. Conclusions: Based on video sequence, the moving object region is detected, and then use DCNN to extract the feature form region based optical flow energy image. Finally, the SVM classifier is used to classify and recognize behaviors. The scheme combines the space-time information of video data, avoids the preprocessing operation of video sequence. It has the characteristics of rapid detection and high accuracy, and it is robust in complex environment. Acknowledgements: This work was supported by the National Social Science Foundation Project (Grants No. 17BSH135). 1Department of Kinesiology, Shenyang Sport University, Shenyang, China; 2School of Science, Northeastern University, Shenyang, China Objectives: In order to find out the gait characteristics for different age people when they walking in different speed, this paper studied the kinematics and kinetics characteristics of the lower limbs in different age groups during walking. And try to find out the causes of injury in the lower limbs of walking. Methods: 45 male and 45 female participants whose arch is normal performed barefoot walking in controlled speed (normal = 1.1m/s, medium = 1.4m/s, and fast = 1.7m/s). They were divided into 3 groups based on the age, age from 30 to 39, 40 to 49, 50 to 59years old, in which there are 15 males and 15 females respectively. The synchronized motion and ground reaction force (GRF) data were captured by Codamotion capture system, ATMI force platform and Foot scan. Results: The results showed that with the grown of age, the percentage of Stance time was long. And the maximum value was 66.78 %, which appeared in the left foot of 50 - 59years old male. The peak value of the first contact GRF in vertical direction was 1.62 BW (normalized with body mass), which happened on the left foot of 50-59years old women, corresponding time is 23.74 %. In all groups, the male step length is greater than the female\'s, and the difference is significant. There was bigger displacement in side direction in the middle of the foot between 50 to 59years old people, which result it is more difficult to keep stability for them. The peak force of first contact in the oldest group (50-59years old) was significantly greater than that of younger group (20-30years old). While the peak stretch force was significantly less than that of younger group (20-30years old). Conclusions: The peak plantar pressure was significantly different in two groups, which was related to the age. Increasing speed was associated with higher injury frequency. That is because with the growing of walking speed, stance time was shorter and the vertical and anterior-posterior GRF were increased during impacts. And the angel of knee extension and ankle joint flexion was larger at high walking speed compared with normal speed. In addition, increased knee extension angle during toe-off was found in high speed, which result in the GRF became lager with the growing of speed. Acknowledgements: This work was supported by the Public Welfare Research Fund of department of Science and Technology of Liaoning (Grant No. 20170027). Objectives: The college financing is one of the most important aspects in college management, and also a fundamental basic link of college operation. As the economy and society of our country develops, the implement environment of College financing management is in a state which experienced continuous updates changes. In this condition, how to through scientific management to make college financing management more adaptable to specific financial work has become a significant aspect. Methods: To strengthen the management of college financial, explore the robust management mode of college financial, corresponding strategy must be used. First, a healthy management philosophy should be innovated. With the increasing diversification of the funding channels of colleges, how to integration, coordinate and optimize the resource of colleges is also a significant sign of measure the financial management ability of colleges. Second, the content of college financial management should be innovated. The goal of financial management is to maximize the interests of stakeholders, including the government, the running scholars, teachers and students and other subjects. Among these interest subjects, college should under the instruction of \"people first”, and give priority to the requirement of teachers and students. At the meantime, the profit for several sides should be coordinated and the correlations between college and society should be well treated. Through the selection of financial management objectives, social benefits as well as the economic benefits will be simultaneously achieved. Results: With the continuous reform and improvement of China\'s social and economic system and the guidance of current scientific development, the main responsibilities of colleges have certain changed. College should own certain foundation of economics and good management tools to support the corresponding development. Conclusions: It is an important task to find a suitable financial management mode for financial management of colleges in the current economic environment. Based on the analysis of fundamental forms which college financial management work faced, several strategies on financial management concept, management content and management mechanism were proposed, which hope to act as a reference to administrative staffs for college financial management. Acknowledgements: This work was supported by the Science and Technology Plan Projects of Shaanxi Provincial Department of Education (Grants No. 18JK0934). Objectives: Private Colleges are obtaining more convenient development and opportunities while facing problems. Capital shortage has become one of urgent problems needs to be solved. Private colleges is a kind of educational institution which established by non-state financial fund. Contribution is far from being desired depending on tuition fee income or donation from organizers. It needs to combine practical development to find more reasonable financing channels, and provide more life blood to college healthy development. Methods: Private Colleges should combine the practical condition of its development and explore multiple financing channels, thus tamping developing keynote of college development. The first is enhancing cooperation and communication with social enterprise. In the condition of enterprise becomes shareholder of college, and rational shareholder reform, the problem of capital shortage can be eased. The second is attempting to publish educational stocks and obtain more capital. With regulations of company law, shares can be issued. In the condition of shareholding reform, Logistics Company will bring to the shareholder team. The last is developing self-study and entering into educational training market by applying material advantages of resource and human power in campus. More income can be obtained on the basis of improving sales service, facilities rental service and research service. Results: Private College is an important power in education system. It provides many fundamental and technical talents, so it is irreplaceable in the area of promoting education spreading. With development of era and progress of society, private colleges in China are faced more challenges and pressure, especially the problem of capital shortage. In the newest situation, how to explore more health financing channels and obtain sustainable development patterns have become significant and considerable. Conclusions: Financing work is a long-term, systematic project. We need to summarize practical experience and learn essence from others, and explore more new mode and methods of financing channels which are beneficial to Private College. Acknowledgements: This work was supported by the Science and Technology Plan Projects of Shaanxi Provincial Department of Education (Grants No. 18JK0930). Multiple coupling model research of the eastern fringe of the Tibetan plateau\'s economic development in the national ecological function zones Lixue Yang Southwest Minzu University, Chengdu, Sichuan, 610041, China Objectives: The ecological niche of the Tibetan Plateau in China and even in the world is determined by its unique natural environmental features. As is specified in the National Principal Function Planning, the Tibetan Plateau is seen as the biggest ecological barrier in China, which has direct impact on the formation and evolution of the monsoon climate and formed a treasure and unique alpine ecosystem. It is also the waterhead of major rivers in Asia and therefore honored as the \"ecological source”, \"climate stream” and \"water tower of Asia”, which shows its critical ecological function and strategic location. Meanwhile, as a national key ecological function area, the advanced economic development model of the developed areas cannot be copied and applied and large-scale resource exploitation is restricted. Hence, in order to curb the vicious cycle of ecological deterioration, economic stagnation and acceleration of social conflicts and to harvest a sound development of both the ecological construction and economic development in a stable, balanced and highly efficient manner, a novel development model successfully coupling ecological protection and economic development must be adopted in the historical context of sustainable development, within the scope of ecological capacity and on the basis of resource endowment and regional features so as to comprehensively prioritize and enhance the systemic function and operational efficacy of the eco-economy, which is also a problem to be addressed in due course in the regional economic development of the Tibetan Plateau area. Methods: By adopting a dynamic panel model and eco-footprint model respectively to discuss the relationship between ecological protections, economic development and conducting a quantitative analysis of the status quo of the eco-economy system. This paper analyzes the problems hidden behind the ecological construction and economic development and the reasons for development stagnation. Following the procedure of discovering and analyzing question, empirical analysis and solving problem, this paper adopts the rationale and framework of mutual complementarity and inter-dependent development of \"ecology, economy and society” and proposes the following proceedings under the premises of taking advantage of resource endowment and the role of national key eco-function area. Results: The Tibetan Plateau shall follow the technological route of innovation coupling mechanism and adopt the development path of institutional, industrial, commercial and technological innovation while adhering to the tenet of eco-protection, construction, setting eco-industrialization and industrial ecologicalization as its breakthrough point, with an aim of realizing the multi-sharing and multi-win-winsituation of ecological protection and construction, economic sustainable development and the well-off livelihood for farmers and shepherds. This study is of great theoretical and practical significance by providing an optional path of economic development and innovation for the sustainable development in the eastern fringe of the Tibetan Plateau area. Fund: Fundamental Research Funds for the Central Universities, Southwest Minzu University (2017NZYQN41). Analysis of healthy development mode of school cooperated with enterprise from the perspective of innovation concept Kaisheng Zhang, Jianjun Hao, Jiangping Chen, Pan Cheng Shaanxi Fashion Engineering University, Xi`an, 712046 P.R. China Objectives: Based on the concept of \"service as the aim and employment as the guidance”, new demands for talents bring new requirements to the training concept of vocational school, so practice teaching is the focus of the talent training. Vocational education is a binary system, including theory and practice. Schools are good at theoretical teaching while enterprises are good at practical teaching, the cooperation between school and enterprise complete the organic unity perfectly. From the view of innovation, cooperation between school and enterprise is a fundamental approach for vocational education. It follows that researching and analyzing healthy development modes of cooperation between school and enterprise is very important. Methods: The cooperation of the school and the enterprise would complete through the following way: first, an innovative practical teaching platform based on the scientific research projects should be build. Constructing cooperation relations between school and enterprise and improving research works development; second, enterprise experts to develop the new teaching model of the school-enterprise cooperation should be employed. Hiring professionals to supply teacher resources; third, the scope of practical teaching and develop the model of practical teaching should be expanded. Letting students into enterprise operation; last, a practice base outside out the school should be established. Results: School-enterprise cooperation is an important method to cultivate innovative talents and a concrete embodiment of innovative teaching concepts in the university. Give full play to the advantages of both school and enterprise is a great way to realize perfection and healthy development of the cooperation between school and enterprise. Conclusions: This paper gives a view of innovation teaching concept and provides intensive analysis on the advantages of cooperation between school and enterprise, and methods of cultivating innovative talents. Cultivating innovative talents, teaching works and teaching concepts should be innovative. The school-enterprise cooperation based on the practical teaching is a new path of teaching innovation and a new teaching model. Only in this way can the healthy development of school-enterprise cooperation be promoted. The study on cognitive characteristics and attitudinal orientation of anti-doping of youth athletes Chuansheng Dong1, Liyan Wang2, Hongwen Zhao1 1Shenyang Sport University, Liaoning, China; 2College of Rehabilitation Science, Shanghai University of Medicine and Health Science, Shanghai, China Objectives: It is to discuss the impact of youth athletes’ gender, degree of education and level of sport on their cognitive characteristics and attitudinal orientation of anti-doping. Methods: The method of cluster sampling by questionnaire is used to investigate 2254 youth athletes from 16 amateur sports schools in China. 2254 questionnaires are distributed and collected with 1718 valid ones, which is analyzed through independent-sample test and analysis of variance. Results: On cognitive habits of anti-doping, there is a difference (t=5.22, P 0.01) between genders with male athletes (M=3.67, SD=1.04) higher than female athletes (M=3.50, SD=0.99). There is a significant impact of degree of education(F=3.223, P 0.05): the higher the degree is, the greater the cognitive habit is. There is no impact of sports level on the cognitive habits; There is a significant impact of genders on the implicit information of anti-doping (t=8.49, P 0.01) with male athletes (M=3.76, SD=1.00) higher than female athletes (M=3.61, SD=1.01). There is no significant impact of the degree of education and level of sports(P 0.05); There is a significant impact of genders on way of logic thinking (t=4.30, P 0.05) with male athletes (M=3.98, SD=0.92) higher than female athletes (M=3.89, SD=0.82). There is a significant impact of sports level (F=7.70, P 0.01). The compared results show that the higher the sports level is, the stronger the logic thinking of anti-doping is; There is no impact of gender and sports level on the attitudinal orientation (P 0.05). There is a significant impact of degree of education (F=3.70, P 0.05): the higher the degree of education is, the larger the attitudinal orientation of anti-doping is. Conclusions: The cognition of anti-doping of the youth male athletes is better than that of female athletes. The higher the degree of education they get, the better the cognition they have. The cognition is raised with the pace of sports level. The gender and sports level have an impact on the attitudinal orientation of anti-doping. Study on influencing factors of physical education majors’ special selection based on structural equationmodel Huhu Lian1, Cailian Lian2, Jinkang Hou3 1School of Physical Education, Sichuan Agricultural University, Ya\'an, China; 2Xiaoyi Zhenxing Street Primary School, Xiaoyi, China; 3Xiaoyi Yangquan Qu Center School, Xiaoyi, China Objectives: To understand the influencing factors and which factors are the most important influencing factors in the selection of special items for contemporary college students of physical education, so as to provide theoretical basis and reference for guiding students to choose special items correctly. Methods: Delphi method was used to screen and determine the influencing factors of students’ selection of special items, and a questionnaire was made according to the actual situation. 300 students from the School of Physical Education of Sichuan Agricultural University were selected as the samples for investigation. Principal component analysis was used to classify many variables into four principal component factors, based on structural equationmodel set up a comprehensive measurement model for influencing factors when selecting students. Results: Social environment, school environment, family environment and personal factors have a positive impact on students’ choice of special projects, and the degree of influence is in turn: personal factors have the greatest impact, school environment and social environment have the second place, family environment has a slightly smaller impact. Conclusions: Interest and hobby are the most important factors affecting students’ choice of special subjects. School facilities, teachers’ teaching attitudes, curriculum evaluation methods, teachers’ professional and technical level, employment, social acceptance of sports items have a significant positive correlation with students’ choice of special subjects, and family economics. Income and parents’ expectations have a certain impact on student selection. Acknowledgements: The staged achievements of the school – level teaching reform project of Sichuan Agricultural University. Reconstruction and Perfection of Curriculum System of Social Sports Major under the Background of Innovation and Entrepreneurship(X2015046). The staged achievements of the Social Science Union special project of Sichuan Agricultural University. Research on the Reform of Practical Ability Training for PE Majors in Agricultural and Forestry Universities(2017YB10). Financing aid by the key research base of philosophy and social science in Sichuan Province-Sichuan Rural Development Research Center. The Development of Characteristic Sports Projects in Sichuan Rural Areas - Taking Cycling as an Example (CR15l7). 1School of Transportation, Nantong University, Nantong 226019, China; 2Nantong Research Institute for Advanced Communication Technologies, Nantong 226019, China; 3Henan urban planning Institute & Corporation, Zhengzhou 450000, China; 4Traffic Police of Qidong City in China, Nantong 226019, China Objectives: Fatigue driving is one of the main factors affecting traffic safety. They are related with the driver status (age, body mass index, and health etc.) and job characteristics (short or long driving trip). At present, some statistical tests are generally used in existing studies to reveal relationship between these characteristics and degree of fatigue, which is very beneficial to traffic safety. The difference between bus drivers and conventional drivers lies in the fact that bus drivers drive large vehicles to continuously run same trips and a service personnel (service quality is related to passenger flow density), which decides the factors influencing their fatigue of two types of drivers are inconsistent. This paper further describes a novel model to reveal an investigation of factors influencing bus driver\'s fatigue while a classification of bus drivers’ fatigue based on the obtained factors that are related to a sensitivity to fatigue during the long-trip drives. Methods: Our work try to reveal a link between improving bus driving work environment, reasonable training, drivers’ selection and the research of reducing work fatigue based on deep learning theory. Based on these influencing factors obtained from GPS and behavior monitoring video, deep learning theory of bus driving work fatigue influencing factors is built with Matlab 14.0, and its correctness and effectiveness is proved by a case study. Results: The importance of all factors can be quantified for analysis. The five main factors, including traffic conditions, road environments, density of passenger flow, service level and working time, are to play a positive role in bus driver fatigue. Results show that the most important factor is long trips, traffic condition is of secondary importance, and density of passenger flow is the last most important. Based on the warning model using deep learning theory, the cluster analysis of different fatigue degree and their scores of five factors is studied to obtain eight subgroups of bus drivers sensitive to the fatigue resulting from boundary changes of these influence factors. Conclusions: The main contribution of this paper is to find the factors that affect bus driver fatigue. Furthermore, the differences between the factors influencing fatigue of bus drivers and traditional drivers are given. Finally, the warning model of bus driver fatigue is given to reduce traffic safety accidents. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61503201); Natural Science Foundation of the Jiangsu Province in China (BK20161280); the Humanities and Social Sciences Foundation of the Ministry of Education in China (16YJCZH086); Nantong University-Nantong Joint Research Center for Intelligent Information Technology (KFKT2017B08); Six talent peaks project in Jiangsu Province (SZCY-009). Extracellular biological organic matters of sewage sludge treated by biofilter with/without earthworm addition Huiyuan Zhong, Li Zhu, Yang Tian, Hao Wang College of Civil and Architectural Engineering, North China University of Science and Technology, Tangshan, China Objectives: This work attempts to study the degradation and characteristic changes of extracellular biological organic matters (EBOM) in sewage sludge can be figure out by biofilter system with/without earthworm addition. Methods: In this study, two cylindrical biological filter reactor packed with ceramic pellets and naturally ventilated were established, which were named BF1 and BF2. Different from BF1 and BF2, Da Ping Er Hao earthworm was added into BR1. The two reactors operated 60days to test the removal effect of EBOM, simultaneously. The EBOM in the effluent was extracted by NH3·H2O after 60days, and then the EBOM from the sludge was separated into HPO-A, HPO-N, TPI-A, TPI-N and HPI five fractions by Amberlite Resin. Results: On day 60 of BF1 and BF2 system operation, the removal rate of TCOD reached 86.7% and 65.4%, and earthworms in BF1 increased by 34.3%. After 60days treatment, the small molecules of HPI in BF1 accounted 43.6%, larger than 36.4% in BF2, this is mainly due to the addition of earthworms in BF1 to degrade macromolecules more easy. The removal rate of DOC for HPO-A, with the highest molecular weight, up to 96.2% also proves this point. The SUVA value of both systems increased, especially the BF1 reactor, which increased by 32.4%. The results showed that earthworm produces a large amount of aromatic substance in the process of degrading sludge. This conclusion can also be verified by excitation-emission matrix (EEM). The intensity of all EEM peaks in both reactors decreased significantly after 60days operation, but the fluorescence peak of aromatic substance in BF1 decreased least efficiently. Conclusions: Through comparison of the two BF reactors, earthworms can effectively remove macromolecules in EBOM, thus improving the efficiency of biological filter on sludge. Although this process produces some small molecules aromatic substance. Acknowledgements: The authors gratefully acknowledge funding from Projects 17273802D supported by Science and Technology Planning Project of Hebei Province, China. 1School of Physical Education, Sichuan Agricultural University, Ya\'an, China2Xiaoyi Zhenxing Street Primary School, Xiaoyi, China; 3Xiaoyi Yangquan Qu Center School, Xiaoyi, China Objectives To find out whether there is a certain relationship between college students’ BMI and physical performance, especially for obese students; which one is the strongest correlation between BMI and physical fitness or physical function in physical test. Find out the rules and guide students to do scientific exercises. Methods: The basic information of 800 college students was obtained by questionnaires, and the BMI of each student was calculated. The total score of these students’ physical performance in the current semester was collected, and the BMI and their respective physical performance were analyzed by Linear Regression. Results: According to Chinese standard, BMI can be divided into the following categories: 18.5 belongs to lean, 18.5-23.9 belongs to normal, 24 belongs to overweight, 24-27.9 belongs to overweight, 28 belongs to obesity. Research shows that there is a certain relationship between BMI and students’ sports performance; when BMI is between 18.5 and 23.9, it has no obvious relationship with students’ sports performance, but when BMI is more than 24 or less than 18.5, there is a linear relationship between BMI and sports performance, indicating that BMI and sports performance are highly correlated. When BMI 24, the greater the BMI, the worse the endurance quality, and the difference was significant (P 0.05). When BMI 18.5, the smaller the BMI, the worse the strength quality, and the difference was significant (P 0.05). Conclusions: BMI in a normal range should be a direction for students to exercise. Students with different BMI should consciously choose their own weak links to exercise targeted projects. More obese students should do more endurance exercises, and thinner students should strengthen their own strength. Acknowledgements: The staged achievements of the school – level teaching reform project of Sichuan Agricultural University. Reconstruction and Perfection of Curriculum System of Social Sports Major under the Background of Innovation and Entrepreneurship(X2015046). The staged achievements of the Social Science Union special project of Sichuan Agricultural University. Research on the Reform of Practical Ability Training for PE Majors in Agricultural and Forestry Universities(2017YB10). Financing aid by the key research base of philosophy and social science in Sichuan Province-Sichuan Rural Development Research Center. The Development of Characteristic Sports Projects in Sichuan Rural Areas - Taking Cycling as an Example (CR15l7). 1School of Transportation, Nantong University, Nantong 226019, China; 2Nantong Research Institute for Advanced Communication Technologies, Nantong 226019, China; 3Traffic police of Qidong City in China, Nantong 226019, China; 4Henan Urban Planning Institute & Corporation, Zhengzhou 450000, China Objectives: Air pollutants, especially PM2.5, is one aspect of main environmental problems in China. The pollutant sources are related with traditional cook or wood-burning stoves, the energy industry, as well as dominate emissions. However, transport is the major source of emissions in urban areas. Hence, traffic air pollution is also affecting the public health. The main contribution of this paper is to present a new model to reveal impact of traffic-related air pollution on health of pregnancy complications and childhood illness using a government emission -health database. Methods: Our work try to reveal how air pollution affects pregnancy complications and childhood illness based on deep learning theory. The related inputs includes average monthly concentrations of traffic air pollutants (i.e., PM2.5 and Nox/NO2) as well as basic information of pregnant women and childhood, including age, body weight, medical history, etc. Results: Results show that mothers exposed to heavy traffic air pollutants more easily give birth to underweight or smaller babies. Further, higher levels of PM2.5 mainly increases 2.7% to 6.8% low-birth-weight babies and 1.2% to 3.4% babies of being small for gestational age. When cutting the average concentration of traffic air pollutants by just 10.7% could reduce 3.2% low-birth-weight babies in each year. Conclusions: The findings of the study is basically consistent with Ryan O\'Hare\'s study work in the UK, which proves that an urgent need for environmental legislation to reduce traffic-related air pollutions to improve the health of residents in urban areas. Acknowledgements: This work was supported by the National Natural Science Foundation of China (61503201); Natural Science Foundation of the Jiangsu Province in China (BK20161280); the Humanities and Social Sciences Foundation of the Ministry of Education in China (16YJCZH086); the open fund for the Key Laboratory for traffic and transportation security of Jiangsu Province (TTS2016-01); Nantong University-Nantong Joint Research Center for Intelligent Information Technology (KFKT2017B08);Six talent peaks project in Jiangsu Province (SZCY-009). College of Civil and Architectural Engineering, North China University of Science and Technology, Tangshan, P.R. China Introduction: Constructed wetland systems have been effectively applied to treat contaminated water, as it could achieve high removal efficiency with simple construction. Compared with conventional wastewater treatment techniques, constructed wetland was more economical for operating cost. Constructed wetland consisted of substrates, plants and microorganism, and obtained efficient purification by the breakdown of microorganism, filtering, adsorption and sedimentation. However, the low temperature during winter weakened its treatment efficiency. Water source heat pump could adopt electric energy to transform low-grade heat energy into high-grade heat energy. And the heat pump possessed some advantages, including low requirements, high efficiency, energy saving and environmental protection. Therefore, it could supply thermal power to the combined system. Methods: In this study, a concrete constructed wetland with an outlet and sampling ports was constructed. The insulation layer was laid at the bottom and around the constructed wetland, and the heat pump unit was adjacent to the constructed wetland. The circulating heating tubes were led out and arranged in a roundabout manner. The substrates from the bottom to the top were gravels (diameter of 8-12mm), steel slags (diameter of 3-8mm), and zeolites (diameter of 1-3mm). Circulating heating tubes were laid between the two layers of substrates. The combined system of heat pump and constructed wetland could make heat pump warm up the effluent from the constructed wetland, and returned warmed effluent to the circulating heating tubes. As a result, the temperature of sewage and substrate in constructed wetland would be increased. The combined system was tested for its ability to remove COD, NH3-N, TN and TP from the contaminated of wastewater treatment plant. Meanwhile, the performance of removal rate and kinetic analysis in the system were discussed. Results: With the increased of HRT, TP removal rate increased at first but later fell down sharply. Meanwhile, TN removal rate increased quickly and significantly. Then, the removal efficiency was slightly reduced, but the curve showed an upward but modestly trend. The removal efficiency of NH3-N reduced over time, but maintained the value above 90%. The trend of COD removal rate was similar to the TN removal rate. Increasing pH could lead to a downtrend on removal efficiency of pollutants. And the pH varied from 9 to 10 had driven the treatment toward better removal result. In addition, the steel slag had better removal performance than gravel, and the gravel had better removal performance than zeolite. The relation of equationcorrelation coefficient for removing pollutants model was COD (0.9602) TN (0.9317) NH3-N (0.8959) TP (0.8848). Conclusions: Compared with constructed wetland independently, the combined system of heat pump and constructed wetland could increase the removal rate of pollutants significantly. Acknowledgements: This work was supported by Science and Technology Planning Project of Hebei Province, China (Grants No. 17273802D). School of Artificial Intelligence, Chongqing Technology and Business University, Chongqing China With the integration of global artificial intelligence and innovative technology, the integration of big data, artificial intelligence and health industry is on the rise. Big data and artificial intelligence will fundamentally change the health care model, and a new era of smart health is coming. Objective: Smart medical care is to create a health information regional medical information platform. The goal is to use big data and artificial intelligence to realize the transmission, storage, query, comparison, display and sharing of medical materials, video and audio information, and to provide patients with high quality and fast. Individualized diagnosis and treatment services make medical treatment more precise, and medical procedures are more standardized and efficient. Method: In order to make better use of big data, artificial intelligence, medical services, and health services, it is necessary to build a smart medical standard system. The first is to build a system of intelligent diagnosis and treatment standards, and establish disease identification, treatment standards and process systems. The second is to build a smart medical technology standard system and establish a technical standard system with four key technologies of interconnection, storage, computing and scenarios as the core. The third is to build a smart medical supervision system and establish intelligent disease prediction, intelligent diagnosis and treatment, intelligent health management and intelligent medicine supervision system. Results: At present, smart medical care has realized application scenarios such as virtual assistant, medical imaging, auxiliary diagnosis, disease risk prediction, drug mining, health management, hospital management, auxiliary medicine, and research platform. Internet of things technology, artificial intelligence, big data computing, etc., have been applied to the discovery, optimization, synthesis, clinical research and other processes of drugs. In February of this year, the US Food and Drug Administration approved the use of CTA imaging to assist in the diagnosis of large blood vessel embolism stroke diagnosis, intelligent medical standards are being explored and established. Conclusion: With the development of mobile Internet, medical care will be developed in the direction of personalization and mobility. In 2015, more than 50% of mobile phone users use mobile medical applications to effectively measure and transmit health data with smart handheld terminals and sensors. From the perspective of development, intelligent medical care focuses on application and development in four aspects: intelligent diagnosis, intelligent treatment, intelligent health management, and intelligent medical supervision. With the development of smart medical care, countries will set up intelligent medical professional committees, which will promote the establishment of industry standards and regulatory systems for smart medical care. Acknowledgements: Chongqing Teaching Reform Project (No. 183103) \"Research and Practice of Artificial Intelligence Talents Training Model”. Influence of whole-body vibration training and stopping training on bone mineral density of old women\'s proximal femur Xinwen Li¹1, Rong Xie¹1, Xiaotang Peng¹1, Liang Cheng²2, Christine A. Rochester³3 1Department of Sports, University of Electronic Science and Technology, Chengdu, China; 2Sports Technology of Sichuan Institute, Chengdu, China; 3Department of Exercise Science, Health and Recreation, Colorado State University -Pueblo, Pueblo, USA Objective: To investigate the influence of 24-week whole-body vibration training and the discontinuance of training for 8-week on bone mineral density of old women\'s proximal femur, and to provide reference for the prevention of osteoporosis in the elderly. Methods: 45 healthy old women aged 58 ˜ 70years old are randomly divided into vibration group (n=22) and control group (n=23). A Power-Platevibrator is applied to conduct load-free whole-body vibration training for 24weeks (3 times/week, 20minutes/time, frequency: 40Hz, amplitude: 3mm), to test the proximal femur BMD of all subjects at 0week, 24weeks and 32weeks. Results: (1) After 24weeks, the BMD of vibration group\'s trochanter and Ward\'s triangular area increase by 12.9% and 13.6% respectively (p 0.05), which are respectively greater than 11.1% and 15.5% of control group (p 0.05); (2) Compared with that of 0week, after stopping training for 8weeks, the BMD of vibration group\'s trochanter increases by 11.3% (p 0.05); After stopping training for 8weeks, the BMD of vibration group\'s trochanter is greater than 13.1% of control group (p 0.05). Conclusion: 40Hz whole-body vibration training can improve the BMD of old women\'s proximal femur, but has little influence on the BMD of femoral neck, and still plays a role in maintaining the increase of trochanter BMD after 8-week training stop. The study finds that WBVT changes the hemodynamics of the subject\'s leg, produces heavy pressure, increases peripheral lymphatic and venous drainage, peripheral and systemic blood flow, causes increased bone flow perfusion, and thus changing the BMD. Acknowledgements: Supported by the Fundamental Research Funds for the Central Universities of China: Physical Education and Health Promotion (No. ZYGX2018007 FRJH). Ultrafast camera with MHz repetition rate for 2D time-resolved fluorescence imaging of cells and tissues Wen Ye1, Yang Yang2, Peng Xu2, Bo Wang2, Bingli Zhu2, Weiwei Cao2, Yonglin Bai2 1Department of Pediatric Neurology, Northwest Women\'s and Children\'s Hospital, Xi\'an, China; 2Key Laboratory of Ultra-fast Photoelectric Diagnostics Technology, Xi\'an Institute of Optics and Precision Mechanics of Chinese Academy of Sciences, Xi\'an, China Objectives: Cancer and normal tissues have different behaviors of auto-fluorescence lifetime which are in the range of several nanosecond. Brain activation leads to cerebral hemodynamic changes will result in variations of the time profiles of detected photons, which will last from 400 to 1200 picosecond. Therefore, detector with 2D imaging capability and temporal resolution up to hundreds of picosecond is critical for diagnosis and locating cancer tissues and for brain activation studies. The existing ultrafast imaging technologies such as streak camera and time-correlated single photon counting (TCSPC) can achieve such high temporal resolution. However, streak camera can\'t offer 2D images, and TCSPC requires scanning pixel by pixel which is time consuming. In this study, we present an optical imaging apparatus based on an ultrafast intensified CCD camera with 200 picosecond gating and repetition rate up to MHz to directly acquire 2D time-resolved fluorescence images of tissues. Methods: vThe proposed optical imaging apparatus consists with a picosecond laser as light source, an ultrafast time-gated intensified camera with MHz repetition rate, optical modules such as microscope and oil immersion objective, and other electronic control units. The tested tissue samples were mixtures of cancer and normal lung tissues that were cut from the edge of tumor tissue and placed on glass slides. The auto-fluorescence lifetime of cancer and normal parts of the samples were measured with a synchroscan streak camera. These diagnostic methods are non-contact and non-invasive. After optical measurements, these tissue were stained by H&E for pathological examination to validate the accuracy of optical diagnostic measurements. Results: The tissue samples were measured by the proposed optical imaging apparatus at different gate time and repetition rates. Each image shows structural changes representing the difference in fluorescence lifetime, which can differentiate lung tumor from healthy tissue. The optical diagnostic results agree with the pathological examinations. According to the synchroscan streak camera measurements, the fluorescence lifetime of cancer tissues (1.2ns on average) is consistently lower than that of normal tissues (3.1ns). Setting the gate time at 1.8ns and repetition rate 2MHz, the acquired image achieve the optimum signal-noise ratio. Conclusions: The feasibility of 2D time-resolved fluorescence imaging using ultrafast camera with MHz repetition rate for cancer diagnosis and localization is verified. It can be a rapid, accurate and non-invasive technique. By switching suitable light source and optical modules, this ultrafast camera can be a potential apparatus for other clinical application such as time-resolved brain functional imaging. Objectives: Catalase is a common enzyme which found in nearly all living organisms exposed to oxygen air environment. Its function is to catalyze the decomposition of hydrogen peroxide to water and oxygen. To study the aging of cell peroxidation due to the lack of catalase due to biological factors, we have learned from many works of literature that nematodes have so-called aging genes that can affect whole organisms without affecting their normal breeding. We chose one of the aging genes DAF-16 single-gene mutant strain nematode in the experiments, and the RNAi technology was used to block the nucleolar regulatory gene NCL-1 indirectly. No previous investigations have done, so we have designed a series of experiments to understand DAF-16 and NCL-1 for aging effect on nematode growth. Methods: Since the 21st century, the direct application of Caenorhabditis elegans has won three Nobel prizes including 2002 in Brenner\'s study of apoptosis, 2006 in Medicine, and 2008 in Chemistry. We wish to understand the hydrogen peroxide accumulation caused by the aging gene DAF-16 with the increase in the metabolic rate caused by the nucleolar regulatory gene NCL-1. The wild-type nematodes obtained from the laboratory and their NCL-1 genes were blocked by RNAi technology to observe their nucleolar formation, growth cycle, and longevity. The aging gene DAF-16 affects the upstream and downstream pathways of the insulin-producing protein of nematodes, leading to indirect effects on spawning and survival, while DAF-2 is upstream of DAF-16 because the DAF-2 mutation suppresses the performance of DAF-16. It has also known that DAF-16 located upstream of the catalase gene. If the gene is mutated, the catalase gene will not express, and the produced catalase will be insufficient to decompose. Hydrogen peroxide accumulated in the body. Researchers have found that about 200 genes affect the regulation of DAF-16, many of which are related to genes in humans. Results: In the nucleolar comparison result of experiments, we can see that the nucleus will appear earlier once NCL-1 blocked, regardless of whether or not DAF-16 is present. In the following experiments, we investigated the survival of these two genes on nematodes. It found that the wild-type nematode had the most extended lifespan, while the DAF-16 and NCL-1 double-mutant nematodes had the shortest lifespan, indicating that when the two genes were not simultaneously expressed, significantly reduced the lifespan of the nematode. Moreover, the expression of either DAF-16 or NCL-1 is decreased slightly, indicating that a single mutation causes the lifespan of the nematode to decrease equally, and the effects of these two genes on the lifespan of the nematode appear to be additive. Conclusions: This results support that the aging gene DAF-16 and the nucleolar regulatory gene NCL-1 mutation, the excessive accumulation of hydrogen peroxide and the early appearance of nucleoli and large nucleoli will affect the length of life of nematodes and the interaction between them. Protective effects of Forsythia suspensa leaves total glycoside on the inflammatory injury of repeated cerebral ischemia-reperfusion model in mice Liling Xiang, Can Wang, Mingsan Miao, Yu Feng, Ming Bai Department Name of Pharmacology, Henan University of Chinese Medicine, Zhengzhou, China Objectives: Study on the Protective effects of Forsythia suspensa leaves total glycoside (FSLTG) on the inflammatory injury of repeated Cerebral Ischemia reperfusion Model in mice. Methods: 90 SPF, mice were randomly divided into sham operation group, model group, Ginado group, Forsythia suspensa leaves total glycoside high, medium and low dose groups. By blocking 15 min of bilateral carotid arteries in mice,then, recovery irrigation 15min, blocking 15min again, Then restore the irrigation again, Establishing repeated cerebral ischemia-reperfusion model in mice, except sham operation group. Before operation, Ginado group(40mg/kg), Forsythia suspensa leaves total glycoside high, medium and low dose groups(400mg/kg, 200mg/kg, 100mg/kg) were given separately to the stomach, sham operation group, model group the same volume of 0.5% CMC, 1 times a day, continuous 7d. Start operation after the last administration of 1h, Postoperative reperfusion 22h, detected serum NSE and S100β levels, detected NO, NOS, ICAM-1, VCAM-1, TNF-α and IL-1β inflammatory factors in brain tissue, observed pathological changes of brain tissue under light microscope. Results: Successfully established repeated cerebral ischemia-reperfusion model in mice. Compared with model group, Forsythia suspensa leaves total glycoside dosage groups decreased the serum NSE, S-100β level in different degrees, At the same time, significantly decreased the level of ICAM-1, VCAM-1 in brain homogenate, attenuating the activity of NOS, increase NO content, significantly alleviated the pathological injury in cortical and hippocampal regions of the brain. Conclusions: Forsythia suspensa leaves total glycoside inhibit excessive formation of Adhesion molecules, thus inhibiting the overexpression of inflammatory factors, reduce inflammation occurrence, It has strong anti-inflammatory effect, To reduce the brain tissue injury caused by repeated ischemia and reperfusion, and protect the brain tissue. Acknowledgements: This work was supported by the Henan Education Department Project (Grants No.15A360033); The Science and technology innovation team of Zhengzhou city (Grants No.131PCXTD612); National base for international cooperation (Grants No.2016-65). A spatial fusion method of heterogeneous images based on a compensatory fuzzy neural network Fusheng Zhang1, Juntao Gu2, Xiangmin Ren1, Zongfu Jia1 1School of Information Engineering, Harbin University, Harbin, China; 2Department Heilongjiang Network Information Security Office, Harbin, China Objectives: According to the different of absorption or reflection about radiation and high-magnetic field in human tissues, researchers have invented many medical imaging technologies, such as B-ultrasound, CT and MRI. To realize the information fusion of CT and MRI heterogeneous medical images, an image fusion method based on compensation fuzzy and neural network (CFNN) is proposed. It can make full use of the complementarity and redundancy of different medical images in describing lesions, so that doctors can master more accurate and comprehensive information of lesion diagnosis and treatment attributes than relying on a single medical image, and then improve the diagnostic rate of doctors. Methods: The medical image fusion algorithm mainly includes the spatial domain method based on the pixels and the clear points of the image blocks. It also includes transform domain method based on wavelet transform and multi-scale geometric analysis. They have the disadvantages of poor contrast, blurred edges and missing details. In order to overcome these problems, this study is to obtain high fidelity image data. Firstly, the original image database is built by convolution sliding window algorithm, and the original heterogeneous data matrix is established. Then, according to the matrix input to RCNN, extract the fuzzy fusion rules or adjust the parameters of the fuzzy fusion rules. Finally, the fusion image is obtained by deconvolution transformation of the fuzzy fusion rules, parameters and heterogeneous data matrix. At the same time, compensation fuzzy technology is used to improve the learning and fusion accuracy of CNN. Results: This method successfully establishes the image data matrix of CT and MRI samples of the human head and chest, and realizes the high-precision fusion of the above images based on CFNN. This algorithm fuses image details, edges and textures with high contrast and good correlation with the original image. Compared with the two traditional algorithms, the fusion effect shows the superiority of the algorithm. Conclusions: In this study, convolution sliding window algorithm is used to establish a heterogeneous medical image fusion algorithm based on fuzzy compensation logic for non-linear feature extraction and dynamic fusion between CNN layers. It points out the features extraction of complex polymorphism and heterogeneous medical images, the acquisition of expectation functions and the cognitive mechanism of intelligent fusion. It is also pointed out that medical image fusion will develop towards multi-modal, dynamic, three-dimensional and hologram. Acknowledgements: This work was supported by the Harbin special Foundation for scientific and technological innovation talents (RC2016JQ009005) and the special fund for discipline construction of Harbin University. 1School of International Business Administration, Shanghai University of Finance and Economics, Shanghai, China; 2Shanghai Chenguan Dairy Co., LTD, Shanghai, China; 3School of Physical Education, Huaqiao University, Quanzhou, China4Department of Physical Education, Wuhan University of Engineering Science, Wuhan, China Background: China is a country with the most developed network consumption and network consumption has become a part of Chinese people\'s life. Wanghong is new celebrities in network age. Wanghong Economy is the outcome of development of network consumption. Wanghong is the core competence of Wanghong Economy and it has become the force of development of network consumption. Meanwhile, Wanghong has influenced the network consumption attitude and changed the consumption custom of consumer in China. The aim of research is discuss the type of network consumption attitude base on the dualistic consumption theory and Wahong Economy in China. Methods: This article uses the Hedonic/Utilitarian Scale survey 1177 people come from 10 provinces in China. And the scale is developed by Spangenberg and Grohmann at 2003. There are 688 female and 489 male consumers are surveyed. The survey reliability coefficient of whole scale is 0.93, and reliability coefficients of 4 dimensions are 0.92, 0.90, 0.93 and 0.91. Results: (1) The score of female consumers’ hedonic consumption attitude is significant difference from male when they buy physical goods at network (t=3.454, p=0.001 0.01**). (2) The score of 20-39years consumers’ hedonic consumption attitude is significant difference from 40-49years consumers when they buy physical goods and service goods at network (p=0.000 0.01**). (3) Different group of profession (p=0.02 0.05*) and economy conditions (p=0.00 0.01**) have different network consumption attitude. (4) The score of consumers’ consumption attitude have significant difference (p=0.04 0.05*) when they buy physical goods and service goods, and the score of hedonic attitude is higher than utilitarian attitude. Conclusions: (1) Female consumers are enjoying in the consumption process in network consumption and female consumers are easier attracted by Wanghong than male. (2) Network consumption is promoting the development of hedonic consumption attitude and Wanghong has pushed the funs network economy in China. (3) The consumer group of 40-49years have specific consumption attitude when they buy something at network and the consumer group of 20-39years is enjoy in the process of consumption at network in China. (4) Profession and economy conditions are the main factors that influence network consumption attitude in China. (5) Hedonic consumption attitude would be main tendency that Chinese people network consumption in future. The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username