Mental Stress Detection System for Desk Workers using Multi-modal Sensor Data

• Item type: SIG Technical Reports(1)
• 公開日: 2024-03-05
• タイトル: Mental Stress Detection System for Desk Workers using Multi-modal Sensor Data
• タイトル / 言語: en
• タイトル: Mental Stress Detection System for Desk Workers using Multi-modal Sensor Data
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: IA-E
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者所属: Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者所属: Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者所属(英): en
• 著者所属(英): Graduate School of Information Science and Engineering, Ritsumeikan University
• 著者名: Zhenghan, Li; Hideaki, Miyaji; Hiroshi, Yamamoto
• 著者名(英): Zhenghan, Li; Hideaki, Miyaji; Hiroshi, Yamamoto
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: As the aging of society intensifies, the workload of desk workers gradually increases, leading to many of them experiencing overwork and causing heightened job-related stress. However, they rarely pay attention to their health problems, and do not go to the hospital for health check-ups due to their heavy work schedules. As a result, many desk workers suffer from various health problems and even have the probability of overwork death due to prolonged work stress. To address these issues, existing researches attempt to use medical equipment and cameras to conduct real-time health measurements for desk workers, which results in high costs and privacy invasion issues. Therefore, we propose a new system that uses low-cost, non-contact multi-sensors to measure the vital indicators of desk workers in real-time. Specifically, the proposed system utilizes infrared sensors to monitor the body and environment temperature. In addition, the sitting posture of desk workers is identified by analyzing the temperature images. At the same time, we employ millimeter wave sensors (in other words, mm-wave sensors) to monitor the heart and respiration rates of desk workers. Then, we apply a multi-modal deep learning model based on an attention mechanism to extract vital features from various vital indicators. This model can determine the different influences of various vital features on health through the attention mechanism, thereby determining a person’s health status and mental stress level. The system sends the determined results to desk workers in real-time through their terminal devices and offers them advice (e.g., take a break or go to the hospital) when potential health problems are detected.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: As the aging of society intensifies, the workload of desk workers gradually increases, leading to many of them experiencing overwork and causing heightened job-related stress. However, they rarely pay attention to their health problems, and do not go to the hospital for health check-ups due to their heavy work schedules. As a result, many desk workers suffer from various health problems and even have the probability of overwork death due to prolonged work stress. To address these issues, existing researches attempt to use medical equipment and cameras to conduct real-time health measurements for desk workers, which results in high costs and privacy invasion issues. Therefore, we propose a new system that uses low-cost, non-contact multi-sensors to measure the vital indicators of desk workers in real-time. Specifically, the proposed system utilizes infrared sensors to monitor the body and environment temperature. In addition, the sitting posture of desk workers is identified by analyzing the temperature images. At the same time, we employ millimeter wave sensors (in other words, mm-wave sensors) to monitor the heart and respiration rates of desk workers. Then, we apply a multi-modal deep learning model based on an attention mechanism to extract vital features from various vital indicators. This model can determine the different influences of various vital features on health through the attention mechanism, thereby determining a person’s health status and mental stress level. The system sends the determined results to desk workers in real-time through their terminal devices and offers them advice (e.g., take a break or go to the hospital) when potential health problems are detected.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA12326962
• 書誌情報: 研究報告インターネットと運用技術（IOT）; 巻 2024-IOT-64, 号 33, p. 1-7, 発行日 2024-03-05
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8787
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会