Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336395
Eito Suda, H. Iba
In recent years, many different areas of research have utilized neural networks (NNs). Many have investigated weights in NNs as well as structural optimization of NNs via neural architecture search. In this paper, we apply weight training during neural architecture search by Ant Colony Optimization(ACO) to two problems from OpenAI Gym and one problem from pybullet-gym controlled by NNs. We also compare the timing of when the weight training is performed, which is before the architecture search, during architecture search and after architecture search. It was found that performing architecture search by ACO and weight training simultaneously is effective for increasing the score of NNs and that by performing weight training before or at the same time as architecture search, the score was increased statistically significantly for all problems compared with fully-connected NN and the score by performing weight training after architecture search was increased statistically significantly for only one problem from pybullet-gym.
{"title":"Neural architecture search and weight adjustment by means of Ant Colony Optimization","authors":"Eito Suda, H. Iba","doi":"10.1109/ICIIBMS50712.2020.9336395","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336395","url":null,"abstract":"In recent years, many different areas of research have utilized neural networks (NNs). Many have investigated weights in NNs as well as structural optimization of NNs via neural architecture search. In this paper, we apply weight training during neural architecture search by Ant Colony Optimization(ACO) to two problems from OpenAI Gym and one problem from pybullet-gym controlled by NNs. We also compare the timing of when the weight training is performed, which is before the architecture search, during architecture search and after architecture search. It was found that performing architecture search by ACO and weight training simultaneously is effective for increasing the score of NNs and that by performing weight training before or at the same time as architecture search, the score was increased statistically significantly for all problems compared with fully-connected NN and the score by performing weight training after architecture search was increased statistically significantly for only one problem from pybullet-gym.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125560963","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336410
Itto Uehara, H. Higa
This paper describes a computer assisted surgical educational system using VR (virtual reality). We made a model to represent bleeding and 3D anatomical human body model to be observed using Unity virtual space. Three able-bodied experimental subjects tested the system with each model. From the experimental results of the bleeding model, we were able to confirm that the blood comes from a cut when we touch any part of the model with a scalpel, surgical knife. From those of the human body model, it was found that we can observe the 3D anatomical human body model from various angles in VR development environment.
{"title":"Computer Assisted Surgical Educational System Using Virtual Reality","authors":"Itto Uehara, H. Higa","doi":"10.1109/ICIIBMS50712.2020.9336410","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336410","url":null,"abstract":"This paper describes a computer assisted surgical educational system using VR (virtual reality). We made a model to represent bleeding and 3D anatomical human body model to be observed using Unity virtual space. Three able-bodied experimental subjects tested the system with each model. From the experimental results of the bleeding model, we were able to confirm that the blood comes from a cut when we touch any part of the model with a scalpel, surgical knife. From those of the human body model, it was found that we can observe the 3D anatomical human body model from various angles in VR development environment.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127299675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336428
Ying Chen, Jineng Li, Dongzhen Wang
Multi-view remote sensing registration has important applications in ground target recognition, image aided navigation, missile image guidance and so on. In order to improve the registration accuracy of remote sensing image with change of view angle, a registration method combining attention mechanism and residual network is proposed. The residual network is used as the backbone structure of feature extraction to improve the abstract ability of the model for complex features. At the same time, the attention mechanism based on channel and spatial is introduced into the feature extraction network to improve the distinguish and location ability of the model to image features. Finally, in the feature matching stage, a mutual correlation operation that improve the performance of feature matching is proposed. Compared with the comparison method, the registration accuracy is improved by 10% on average, Experiments show that this method improves the accuracy of multi view remote sensing image registration.
{"title":"Remote Sensing Image Registration based on Attention and Residual Network","authors":"Ying Chen, Jineng Li, Dongzhen Wang","doi":"10.1109/ICIIBMS50712.2020.9336428","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336428","url":null,"abstract":"Multi-view remote sensing registration has important applications in ground target recognition, image aided navigation, missile image guidance and so on. In order to improve the registration accuracy of remote sensing image with change of view angle, a registration method combining attention mechanism and residual network is proposed. The residual network is used as the backbone structure of feature extraction to improve the abstract ability of the model for complex features. At the same time, the attention mechanism based on channel and spatial is introduced into the feature extraction network to improve the distinguish and location ability of the model to image features. Finally, in the feature matching stage, a mutual correlation operation that improve the performance of feature matching is proposed. Compared with the comparison method, the registration accuracy is improved by 10% on average, Experiments show that this method improves the accuracy of multi view remote sensing image registration.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126458207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336406
Zhixing Jiang, Guangming Lu, Dafan Zhang
In traditional Chinese medicine (TCM), wrist pulse is of great significance to help doctors in diagnosis. With the development of sensing technology, the computerized wrist pulse analysis has been attracting more attention in modern medicine for its non-invasive and convenient. Considering the TCM pulse diagnosis theory, it is necessary to develop effective feature extraction methods for computerized diagnosis. In this paper, we decompose the pressure pulse waveform of the radial artery to several components by sparse decomposition with Gabor function. In order to better represent the pulse waveform signal, we use an Gabor function based on the characteristics of the pulse waveform to generate a time-frequency dictionary. Compared with the conventional representation methods, the shape of the Gabor function is more variable, which can better represent both the contour and specific peaks. In addition, due to the limitation of the windowing, the Gabor function can reduce the influence on other positions when representing specific position. The feature vector composed of the decomposed components can be used for the computerized pulse signal analysis and disease diagnosis. The experimental results show that the proposed method can exhibit superior performance in distinguishing between the signals collected from patients and healthy individuals.
{"title":"Sparse decomposition of pressure pulse wave signal based on time frequency analysis","authors":"Zhixing Jiang, Guangming Lu, Dafan Zhang","doi":"10.1109/ICIIBMS50712.2020.9336406","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336406","url":null,"abstract":"In traditional Chinese medicine (TCM), wrist pulse is of great significance to help doctors in diagnosis. With the development of sensing technology, the computerized wrist pulse analysis has been attracting more attention in modern medicine for its non-invasive and convenient. Considering the TCM pulse diagnosis theory, it is necessary to develop effective feature extraction methods for computerized diagnosis. In this paper, we decompose the pressure pulse waveform of the radial artery to several components by sparse decomposition with Gabor function. In order to better represent the pulse waveform signal, we use an Gabor function based on the characteristics of the pulse waveform to generate a time-frequency dictionary. Compared with the conventional representation methods, the shape of the Gabor function is more variable, which can better represent both the contour and specific peaks. In addition, due to the limitation of the windowing, the Gabor function can reduce the influence on other positions when representing specific position. The feature vector composed of the decomposed components can be used for the computerized pulse signal analysis and disease diagnosis. The experimental results show that the proposed method can exhibit superior performance in distinguishing between the signals collected from patients and healthy individuals.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123741134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336420
Cang Naimeng, Yuan Wanjun
Corner detection is an important basic research content in the fields of image processing and computer vision. Corner detection algorithm based on image contour is widely used in image registration, target recognition and shape analysis. In this paper, the existing corner detection algorithms based on contour are classified in detail, the main ideas of the algorithms are elaborated, the advantages and disadvantages of each algorithm are summarized, and the existing problems and further research directions are pointed out. It is expected to provide reference for the research of corner detection based on image contour.
{"title":"A Review of Corner Detection Algorithms Based on Image Contour","authors":"Cang Naimeng, Yuan Wanjun","doi":"10.1109/ICIIBMS50712.2020.9336420","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336420","url":null,"abstract":"Corner detection is an important basic research content in the fields of image processing and computer vision. Corner detection algorithm based on image contour is widely used in image registration, target recognition and shape analysis. In this paper, the existing corner detection algorithms based on contour are classified in detail, the main ideas of the algorithms are elaborated, the advantages and disadvantages of each algorithm are summarized, and the existing problems and further research directions are pointed out. It is expected to provide reference for the research of corner detection based on image contour.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130129028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336408
Qihui Ai, Bei Wang, Jing Jin, Xingyu Wang
Sleep is important to human beings. A good sleep can bring sufficient energy and keep healthy. Sleep process contains several sleep stages. There are also several short-duration awake states appeared in sleep process. The amount of those awake states during sleep would be related to evaluate the sleep quality. In this study, the automatic detection method of awake state is investigated on the sleep recording of patient with sleep disorder. Two nonlinear features of ApEN and Mc-ApEN are calculated based on the EMG signal recorded during one's overnight sleep process. The detection performance is analyzed and compared. The obtained results showed that the presented method is feasible for awake state detection during sleep, where Mc-ApEN is slightly better than ApEN. It can be an assistant tool for sleep evaluation.
{"title":"Automatic detection of awake state during sleep based on the approximate entropy of EMG","authors":"Qihui Ai, Bei Wang, Jing Jin, Xingyu Wang","doi":"10.1109/ICIIBMS50712.2020.9336408","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336408","url":null,"abstract":"Sleep is important to human beings. A good sleep can bring sufficient energy and keep healthy. Sleep process contains several sleep stages. There are also several short-duration awake states appeared in sleep process. The amount of those awake states during sleep would be related to evaluate the sleep quality. In this study, the automatic detection method of awake state is investigated on the sleep recording of patient with sleep disorder. Two nonlinear features of ApEN and Mc-ApEN are calculated based on the EMG signal recorded during one's overnight sleep process. The detection performance is analyzed and compared. The obtained results showed that the presented method is feasible for awake state detection during sleep, where Mc-ApEN is slightly better than ApEN. It can be an assistant tool for sleep evaluation.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126494476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336415
N. Kurita, Ayano Fujinaka, Kojiro Mekata, H. Takizawa, H. Kudo
In this report, we have proposed a construction method of polar coordinate systems to define the positions of cervical structures in videofluorography (VF) during swallowing. The polar coordinate systems have s and t axes on radial and circumferential directions, respectively, and can be translated and rotated to track cervical spines in VFs. In addition, the t axes can be bent to represent the bending deformation of cervical spines. The parameters of the translation, rotation and bending deformation are obtained by a deformable template matching technique. The accuracy of the proposed coordinate systems was evaluated by use of actual VFs, and several experimental results were shown.
{"title":"Preliminary study on construction of polar coordinate systems to define the positions of cervical structures in videofluorography during swallowing","authors":"N. Kurita, Ayano Fujinaka, Kojiro Mekata, H. Takizawa, H. Kudo","doi":"10.1109/ICIIBMS50712.2020.9336415","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336415","url":null,"abstract":"In this report, we have proposed a construction method of polar coordinate systems to define the positions of cervical structures in videofluorography (VF) during swallowing. The polar coordinate systems have s and t axes on radial and circumferential directions, respectively, and can be translated and rotated to track cervical spines in VFs. In addition, the t axes can be bent to represent the bending deformation of cervical spines. The parameters of the translation, rotation and bending deformation are obtained by a deformable template matching technique. The accuracy of the proposed coordinate systems was evaluated by use of actual VFs, and several experimental results were shown.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134132698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/iciibms50712.2020.9336206
Huang Jin-bo, Yang Rui-jun, Cheng Yan
A combined Q and Sarsa algorithm based on united simulated annealing strategy proposed in order to solve the problems that the convergence speed of traditional reinforcement learning algorithm is slow. It could balance the relationship between trial-error and efficiency among different methods by a random dynamic adjustment factor method. The roulette algorithm is used to improve Q-learning, and the simulated annealing algorithm is used to replace the selection strategy of Sarsa algorithm, and the overall convergence rate of the algorithm is controlled by the annealing rate. Finally, the task of reward function is subdivided, and the reward function based on action decomposition is designed. The simulation results show that the improved path planning method can effectively reduce the time cost and the collision times of the first path finding.
{"title":"Joint Reinforcement Learning Method Based on Roulette Algorithm and Simulated Annealing Strategy","authors":"Huang Jin-bo, Yang Rui-jun, Cheng Yan","doi":"10.1109/iciibms50712.2020.9336206","DOIUrl":"https://doi.org/10.1109/iciibms50712.2020.9336206","url":null,"abstract":"A combined Q and Sarsa algorithm based on united simulated annealing strategy proposed in order to solve the problems that the convergence speed of traditional reinforcement learning algorithm is slow. It could balance the relationship between trial-error and efficiency among different methods by a random dynamic adjustment factor method. The roulette algorithm is used to improve Q-learning, and the simulated annealing algorithm is used to replace the selection strategy of Sarsa algorithm, and the overall convergence rate of the algorithm is controlled by the annealing rate. Finally, the task of reward function is subdivided, and the reward function based on action decomposition is designed. The simulation results show that the improved path planning method can effectively reduce the time cost and the collision times of the first path finding.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134550072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-11-18DOI: 10.1109/ICIIBMS50712.2020.9336418
Yu Tian, Y. Wan-jun, Miao Zhang, M. Zhang, Jun-Yi Tang
In order to improve the efficiency of emergency management and decision-making and command of infectious diseases, infectious disease early warning technologies and systems can be used to monitor the trend of abnormal outbreaks and provide timely and accurate warnings. However, the current infectious disease early warning system still has problems such as insufficient intelligence, poor circulation of key information, and difficulty in distributed collaborative decision-making. Based on the blockchain distributed computing architecture, combined with emerging information technologies such as artificial intelligence, big data, and smart contracts, an infectious disease monitoring and early warning technology and its implementation framework are proposed. The technology can efficiently concentrate multi-party monitoring forces, flexibly integrate multiple early warning methods, build a knowledge integration and smart interconnected distributed collaborative monitoring environment under the premise of security and privacy protection, and use smart contracts as open, fair, software-defined monitoring The agent integrates decision-making, monitoring infectious diseases and automatic early warning in real time, so as to take into account the requirements of early warning accuracy and timeliness, and avoid isolated evidence and decision bias.
{"title":"The application of blockchain technology in the early warning and monitoring of infectious diseases","authors":"Yu Tian, Y. Wan-jun, Miao Zhang, M. Zhang, Jun-Yi Tang","doi":"10.1109/ICIIBMS50712.2020.9336418","DOIUrl":"https://doi.org/10.1109/ICIIBMS50712.2020.9336418","url":null,"abstract":"In order to improve the efficiency of emergency management and decision-making and command of infectious diseases, infectious disease early warning technologies and systems can be used to monitor the trend of abnormal outbreaks and provide timely and accurate warnings. However, the current infectious disease early warning system still has problems such as insufficient intelligence, poor circulation of key information, and difficulty in distributed collaborative decision-making. Based on the blockchain distributed computing architecture, combined with emerging information technologies such as artificial intelligence, big data, and smart contracts, an infectious disease monitoring and early warning technology and its implementation framework are proposed. The technology can efficiently concentrate multi-party monitoring forces, flexibly integrate multiple early warning methods, build a knowledge integration and smart interconnected distributed collaborative monitoring environment under the premise of security and privacy protection, and use smart contracts as open, fair, software-defined monitoring The agent integrates decision-making, monitoring infectious diseases and automatic early warning in real time, so as to take into account the requirements of early warning accuracy and timeliness, and avoid isolated evidence and decision bias.","PeriodicalId":243033,"journal":{"name":"2020 5th International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129889017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：