Pub Date : 2022-10-11DOI: 10.23919/WAC55640.2022.9934247
Yufu Li
Prefabricated buildings have been highly recognized for their advantages such as ease of construction and construction, and the scope of popularity is becoming wider and more numerous, and there are still a large number of building construction plans to use this type of building form. Under such circumstances, what we need to do most is the design of prefabricated buildings, especially through BIM technology application to improve the design quality of prefabricated buildings, to better guide the construction of prefabricated buildings. To ensure the efficiency and quality of building construction, and reasonably control costs. This article first analyzes the advantages of BIM technology in the design of prefabricated buildings, and then mainly introduces the design of prefabricated buildings based on BIM and conducts research and analysis on BIM technology application in building construction.
{"title":"Research on Design and Construction of Prefabricated Buildings Based on BIM Technology","authors":"Yufu Li","doi":"10.23919/WAC55640.2022.9934247","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934247","url":null,"abstract":"Prefabricated buildings have been highly recognized for their advantages such as ease of construction and construction, and the scope of popularity is becoming wider and more numerous, and there are still a large number of building construction plans to use this type of building form. Under such circumstances, what we need to do most is the design of prefabricated buildings, especially through BIM technology application to improve the design quality of prefabricated buildings, to better guide the construction of prefabricated buildings. To ensure the efficiency and quality of building construction, and reasonably control costs. This article first analyzes the advantages of BIM technology in the design of prefabricated buildings, and then mainly introduces the design of prefabricated buildings based on BIM and conducts research and analysis on BIM technology application in building construction.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122231085","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934512
T. Burton, Émélie Séguin, Marc Doumit
Walking Assistive Exoskeletons (WAEs) are wearable devices that can allow individuals with mobility impairments to maintain their autonomy. The growing elderly population has benefited from these devices by receiving assistance at joints where their muscle function has declined. Typically, the primary objective of WAEs has been to reduce the metabolic cost of walking, allowing users to walk for extended periods. However, this strategy does not address the growing concern that seniors are at an increased risk of falling and sustaining severe injuries. Gait and balance disorders are among the most common causes of falls in the elderly. As people age, it is increasingly important to investigate the musculoskeletal changes contributing to frontal plane instability, as mediolateral and postero-lateral falls are highly correlated with severe injuries. Specifically, the hip abductor and adductor muscles are essential in maintaining balance in the frontal plane, yet minimal research has been conducted on the effect of a hip abduction-adduction exoskeleton on elderly stability. This paper outlines gait biomechanics, muscle contributions for mediolateral stability, changes associated with ageing, and strategies to improve balance during elderly gait. Subsequently, a comparison of existing hip WAEs highlights the feasibility of applying this technology to improve elderly gait stability.
{"title":"Study of Hip Exoskeleton Technology for Elderly Stability During Walking","authors":"T. Burton, Émélie Séguin, Marc Doumit","doi":"10.23919/WAC55640.2022.9934512","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934512","url":null,"abstract":"Walking Assistive Exoskeletons (WAEs) are wearable devices that can allow individuals with mobility impairments to maintain their autonomy. The growing elderly population has benefited from these devices by receiving assistance at joints where their muscle function has declined. Typically, the primary objective of WAEs has been to reduce the metabolic cost of walking, allowing users to walk for extended periods. However, this strategy does not address the growing concern that seniors are at an increased risk of falling and sustaining severe injuries. Gait and balance disorders are among the most common causes of falls in the elderly. As people age, it is increasingly important to investigate the musculoskeletal changes contributing to frontal plane instability, as mediolateral and postero-lateral falls are highly correlated with severe injuries. Specifically, the hip abductor and adductor muscles are essential in maintaining balance in the frontal plane, yet minimal research has been conducted on the effect of a hip abduction-adduction exoskeleton on elderly stability. This paper outlines gait biomechanics, muscle contributions for mediolateral stability, changes associated with ageing, and strategies to improve balance during elderly gait. Subsequently, a comparison of existing hip WAEs highlights the feasibility of applying this technology to improve elderly gait stability.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133966010","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934501
Weibin Guo, Lin Shi, Z. Wu
Cloud environment is a large-scale, distributed and complex system. Due to the complex interdependence and call relationship between its functional layers, the efficient operation and maintenance of cloud environment has become a major problem. The main form of daily monitoring KPI data in cloud environment is time series data. The prediction and anomaly detection of these data have always been two hot spots at home and abroad. The algorithm with high prediction accuracy and high anomaly detection accuracy can help us find the potential problems in the cloud environment and stop the loss in time to avoid large losses. Based on the previous relevant research results, this paper takes the data in the cloud environment as the research object, and takes improving the prediction accuracy and anomaly detection accuracy of the data as the research goal, and puts forward efficient and accurate prediction algorithms and anomaly detection algorithms suitable for the data characteristics in the cloud environment.
{"title":"Research on anomaly detection algorithm of time series data in cloud environment","authors":"Weibin Guo, Lin Shi, Z. Wu","doi":"10.23919/WAC55640.2022.9934501","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934501","url":null,"abstract":"Cloud environment is a large-scale, distributed and complex system. Due to the complex interdependence and call relationship between its functional layers, the efficient operation and maintenance of cloud environment has become a major problem. The main form of daily monitoring KPI data in cloud environment is time series data. The prediction and anomaly detection of these data have always been two hot spots at home and abroad. The algorithm with high prediction accuracy and high anomaly detection accuracy can help us find the potential problems in the cloud environment and stop the loss in time to avoid large losses. Based on the previous relevant research results, this paper takes the data in the cloud environment as the research object, and takes improving the prediction accuracy and anomaly detection accuracy of the data as the research goal, and puts forward efficient and accurate prediction algorithms and anomaly detection algorithms suitable for the data characteristics in the cloud environment.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134352984","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934511
F. Mohammadi, M. Saif, M. Ahmadi, B. Shafai
This paper provides a survey of measurable factors affecting the adoption of cybersecurity enhancement strategies in the smart grid. From power systems operators’ point of view, it is crucial to determine to what degree the smart grid can be made resilient against cyber-attacks through the use of efficient resilience enhancement strategies. In the past decade, there have been numerous attempts to improve the resilience of the smart grid against different types of cyber-attacks. This paper mainly focuses on the recently proposed cybersecurity strategies for successfully detecting and identifying False Data Injection (FDI) attacks, and compares and contrasts their accuracy, computational burden, and robustness against external factors, as their main measurable factors. It is challenging to find an all-inclusive solution satisfying all requirements of the smart grid. Hence, a quantitative mean for evaluation of the cyber-attack detection and identification strategies is provided in this paper.
{"title":"A Review of Cyber–Resilient Smart Grid","authors":"F. Mohammadi, M. Saif, M. Ahmadi, B. Shafai","doi":"10.23919/WAC55640.2022.9934511","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934511","url":null,"abstract":"This paper provides a survey of measurable factors affecting the adoption of cybersecurity enhancement strategies in the smart grid. From power systems operators’ point of view, it is crucial to determine to what degree the smart grid can be made resilient against cyber-attacks through the use of efficient resilience enhancement strategies. In the past decade, there have been numerous attempts to improve the resilience of the smart grid against different types of cyber-attacks. This paper mainly focuses on the recently proposed cybersecurity strategies for successfully detecting and identifying False Data Injection (FDI) attacks, and compares and contrasts their accuracy, computational burden, and robustness against external factors, as their main measurable factors. It is challenging to find an all-inclusive solution satisfying all requirements of the smart grid. Hence, a quantitative mean for evaluation of the cyber-attack detection and identification strategies is provided in this paper.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133646653","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934169
Hu Sheng, Weizhi Sun
With the popularization of the Internet and the rapid development of new technologies such as artificial intelligence, cloud computing, and the Internet of Things(ITS), the explosive growth of data traffic puts forward higher requirements for transmission performance. In order to meet the ever-increasing demand for capacity and cope with the ensuing crisis, the application of digital signal processing technology in the communications field is imperative. With the development of high-speed integrated circuits and digital signal processing algorithms and other technologies, high-speed digital signal processing has become increasingly mature, and ITS has become one of the global research hotspots. The country is also strategically advancing the research on ITS. This article mainly uses the experimental analysis method to explore whether the application of digital signal processing technology can provide a new idea for the development of the communication field in the context of ITS. In the experiment, this article analyzes the OPA multiplexing mode demodulation energy and crosstalk energy, and analyzes the changes of the bit error rate under different communication rates and different communication distances. According to the experimental results, OPA multiplexing transmission has a certain degree of stability. The combination of OPA mode multiplexing technology and traditional systems can provide more possibilities for building communication networks with higher capacity systems in the future. As the communication rate and communication distance increase, the bit error rate will increase accordingly. Therefore, this article studies the application of digital signal processing technology in the communication field under the background of ITS, which has huge application potential and research value.
{"title":"Digital Signal Processing Technology in the Communication Field under the Background of the Internet of Things","authors":"Hu Sheng, Weizhi Sun","doi":"10.23919/WAC55640.2022.9934169","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934169","url":null,"abstract":"With the popularization of the Internet and the rapid development of new technologies such as artificial intelligence, cloud computing, and the Internet of Things(ITS), the explosive growth of data traffic puts forward higher requirements for transmission performance. In order to meet the ever-increasing demand for capacity and cope with the ensuing crisis, the application of digital signal processing technology in the communications field is imperative. With the development of high-speed integrated circuits and digital signal processing algorithms and other technologies, high-speed digital signal processing has become increasingly mature, and ITS has become one of the global research hotspots. The country is also strategically advancing the research on ITS. This article mainly uses the experimental analysis method to explore whether the application of digital signal processing technology can provide a new idea for the development of the communication field in the context of ITS. In the experiment, this article analyzes the OPA multiplexing mode demodulation energy and crosstalk energy, and analyzes the changes of the bit error rate under different communication rates and different communication distances. According to the experimental results, OPA multiplexing transmission has a certain degree of stability. The combination of OPA mode multiplexing technology and traditional systems can provide more possibilities for building communication networks with higher capacity systems in the future. As the communication rate and communication distance increase, the bit error rate will increase accordingly. Therefore, this article studies the application of digital signal processing technology in the communication field under the background of ITS, which has huge application potential and research value.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132658424","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934354
Y. Yoshida, Hiroaki Sakamoto, E. Yuda
The widespread use of wearable sensor technology has made it possible to obtain a variety of human biological information. Among them, workout is important for health promotion, and estimation of exercise duration and intensity is a clear and convenient way to understand health status. Therefore, it is desirable to be able to estimate workout efficiently. Many existing wearable sensors can measure the accumulated intensity of aerobic exercise using heart rate or provide a rough estimate. However, the estimation algorithm has not been published, and it is not clear how accurate the workout can actually be detected. In this study, we attempted to detect workout time from biometric data obtained over a long period of time using random forest. The results showed a high estimation, with 0.96 accuracy and 0.92 recall. As a result, workout was considered easy to estimate.
{"title":"Precision of human workout-time detection using Random Forests and Wearable Sensor Data","authors":"Y. Yoshida, Hiroaki Sakamoto, E. Yuda","doi":"10.23919/WAC55640.2022.9934354","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934354","url":null,"abstract":"The widespread use of wearable sensor technology has made it possible to obtain a variety of human biological information. Among them, workout is important for health promotion, and estimation of exercise duration and intensity is a clear and convenient way to understand health status. Therefore, it is desirable to be able to estimate workout efficiently. Many existing wearable sensors can measure the accumulated intensity of aerobic exercise using heart rate or provide a rough estimate. However, the estimation algorithm has not been published, and it is not clear how accurate the workout can actually be detected. In this study, we attempted to detect workout time from biometric data obtained over a long period of time using random forest. The results showed a high estimation, with 0.96 accuracy and 0.92 recall. As a result, workout was considered easy to estimate.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"90 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133818429","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934140
Yanxue Guo, Juan Wang, Xiang Zhang
With the popularization and rapid development of the Internet and BD, our daily lives are gradually brought from the "information age" to the "data age". All walks of life are making full use of the Internet and BD, which makes the human society produce economically and economically. The social lifestyle has undergone tremendous changes, and this transformation will inevitably affect all areas of information dissemination. This article mainly focuses on the research on the commercial news production of online media and its commercial news dissemination model based on the Internet and BD technology. After analyzing the importance and impact of the Internet and BD on online media, the questionnaire is then used. The survey method analyzed the influence of the Internet and BD technology on the commercial news production of online media and its commercial dissemination mode. The results of the questionnaire survey showed that the Internet and BD technology have the main influence on the commercial news production of online media. That is, the path of production has changed, accounting for more than 43% of the total. This is because data analysis technology can be used to mine valuable news, and then the main body of production has changed, from the original single main body to the multiple main body. The impact on the news business communication mode of the online media is mainly the change of the communication channel, which has changed from a single mode to a multi-mode, accounting for more than 42% of the total number of people.
{"title":"News Production and Business Communication Model of Internet Media Based on Big Data Technology","authors":"Yanxue Guo, Juan Wang, Xiang Zhang","doi":"10.23919/WAC55640.2022.9934140","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934140","url":null,"abstract":"With the popularization and rapid development of the Internet and BD, our daily lives are gradually brought from the \"information age\" to the \"data age\". All walks of life are making full use of the Internet and BD, which makes the human society produce economically and economically. The social lifestyle has undergone tremendous changes, and this transformation will inevitably affect all areas of information dissemination. This article mainly focuses on the research on the commercial news production of online media and its commercial news dissemination model based on the Internet and BD technology. After analyzing the importance and impact of the Internet and BD on online media, the questionnaire is then used. The survey method analyzed the influence of the Internet and BD technology on the commercial news production of online media and its commercial dissemination mode. The results of the questionnaire survey showed that the Internet and BD technology have the main influence on the commercial news production of online media. That is, the path of production has changed, accounting for more than 43% of the total. This is because data analysis technology can be used to mine valuable news, and then the main body of production has changed, from the original single main body to the multiple main body. The impact on the news business communication mode of the online media is mainly the change of the communication channel, which has changed from a single mode to a multi-mode, accounting for more than 42% of the total number of people.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115514831","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934344
Yunus Yetis, K. Tehrani, M. Jamshidi
This paper presents a wind speed forecasting method based on machine learning approach applied to microgrids. A small wind turbine is modeled then the forecasting method is developed to estimate the wind speed and the potential of energy production in a wind farm. The accuracy of this method is higher compared to other existing methods in the literature for time series analysis such as artificial neural networks (ANN). This study is focused on a case study with the real data from weather station of Basel in Switzerland. The results obtained are presented and discussed.
{"title":"A Machine Learning Approach for Wind Speed Forecasting in Microgrids","authors":"Yunus Yetis, K. Tehrani, M. Jamshidi","doi":"10.23919/WAC55640.2022.9934344","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934344","url":null,"abstract":"This paper presents a wind speed forecasting method based on machine learning approach applied to microgrids. A small wind turbine is modeled then the forecasting method is developed to estimate the wind speed and the potential of energy production in a wind farm. The accuracy of this method is higher compared to other existing methods in the literature for time series analysis such as artificial neural networks (ANN). This study is focused on a case study with the real data from weather station of Basel in Switzerland. The results obtained are presented and discussed.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115771695","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934292
Bo Yu, Xiaomin Liu, Xinrui Ju, Ye Wan, Yuanyuan Liu
There are many factors affecting the total construction duration of power transmission and transformation project, resulting in low accuracy of construction duration prediction. And the traditional construction duration prediction method has some limitations to a certain extent. The prediction results mainly rely on historical information materials and expert judgment, so it has always been the difficulty of prediction. Aiming at the existing problems, a total construction duration prediction method of power transmission and transformation project based on BP neural network theory is proposed. The theory has high generalization ability and the ability to solve nonlinear influencing factors, which can greatly improve the accuracy of construction duration prediction.
{"title":"Construction Duration Prediction Model of Power Transmission and Transformation Project Based on BP Neural Network","authors":"Bo Yu, Xiaomin Liu, Xinrui Ju, Ye Wan, Yuanyuan Liu","doi":"10.23919/WAC55640.2022.9934292","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934292","url":null,"abstract":"There are many factors affecting the total construction duration of power transmission and transformation project, resulting in low accuracy of construction duration prediction. And the traditional construction duration prediction method has some limitations to a certain extent. The prediction results mainly rely on historical information materials and expert judgment, so it has always been the difficulty of prediction. Aiming at the existing problems, a total construction duration prediction method of power transmission and transformation project based on BP neural network theory is proposed. The theory has high generalization ability and the ability to solve nonlinear influencing factors, which can greatly improve the accuracy of construction duration prediction.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114915713","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 : 2022-10-11DOI: 10.23919/WAC55640.2022.9934249
Yuki Ito, Kento Morita, T. Wakabayashi, H. Shinkoda, Asami Matsumoto, Yukari Noguchi, Masako Shiramizu
In the neonatal intensive care unit (NICU), the preterm infant located in incubator takes various medical care day and night. Unusual environment in NICU may affect neurodevelopment of newborn subject, some researches evaluate the sleep/wake state of subject by visual or using the Actigraph attached on the leg. This paper proposes a sleep/wake status estimation method using video images and convolutional neural network (CNN) to reduce assessment time and improve the reliability. The Brazelton’s criteria evaluates the newborn’s sleep/wake states in six stages, the proposed method performs six-class classification using 3D CNN. In the experiment, we conducted 4 experiments by using original data, two different frame shifting, and using the frame differential. Experimental results using 16 video of 8 subjects showed that the training using original dataset achieved the highest macro-F1 value (0.766) which improves the macro-F1 value (0.765) of our previous result using support vector machine (SVM) and optical flow. Results also suggested that the 3D CNN improves the classification accuracy but the data augmentation using frame shift is not suitable to our dataset.
{"title":"Automatic Estimation of Neonatal Sleep/Wake States in the NICU Using 3D CNN","authors":"Yuki Ito, Kento Morita, T. Wakabayashi, H. Shinkoda, Asami Matsumoto, Yukari Noguchi, Masako Shiramizu","doi":"10.23919/WAC55640.2022.9934249","DOIUrl":"https://doi.org/10.23919/WAC55640.2022.9934249","url":null,"abstract":"In the neonatal intensive care unit (NICU), the preterm infant located in incubator takes various medical care day and night. Unusual environment in NICU may affect neurodevelopment of newborn subject, some researches evaluate the sleep/wake state of subject by visual or using the Actigraph attached on the leg. This paper proposes a sleep/wake status estimation method using video images and convolutional neural network (CNN) to reduce assessment time and improve the reliability. The Brazelton’s criteria evaluates the newborn’s sleep/wake states in six stages, the proposed method performs six-class classification using 3D CNN. In the experiment, we conducted 4 experiments by using original data, two different frame shifting, and using the frame differential. Experimental results using 16 video of 8 subjects showed that the training using original dataset achieved the highest macro-F1 value (0.766) which improves the macro-F1 value (0.765) of our previous result using support vector machine (SVM) and optical flow. Results also suggested that the 3D CNN improves the classification accuracy but the data augmentation using frame shift is not suitable to our dataset.","PeriodicalId":339737,"journal":{"name":"2022 World Automation Congress (WAC)","volume":"218 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122057953","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：