S. Scataglini, Lenie Denteneer, Nele Struyf, S. Truijen
Work-related injuries involving significant physical loads are of growing concern, as these injuries can negatively impact work productivity, physical health and well-being at work, but also have economic costs due to absenteeism. This observational study aims to quantify the ergonomical assessment of dockworkers using wearable devices to reduce the risk of injuries. Two different gestures are analyzed and observed using a wearable mocap system: Task1A securing the lash bars and Task 1B as lifting and locking process. Additional tests and questionnaire were also requested: FMS, HHD, RAND-36, ODI, and TSK questionnaires. As result, it was found a large spread between the maximum and minimum degrees and the range in which they are working among the different dockworkers. Despite the large heterogeneity, we found for task 1B that the wrist was held in extreme positions, and a significant correlation was found between the time and standard deviation of the low back (R = 0.45, p = 0.045). In addition, all participants reported having complaints; 66% had kinesiophobia, 60% were overweight, and the mean score on the FMS was 15. The occurrence of fatigue, complaints, kinesiophobia, a challenging work environment, or an unhealthy lifestyle may lead to altered movement patterns, which could possibly increase the risk of injury. Based on the results, we recommend implementing a warm-up and an individualized prevention program to reduce the risk of injury.
涉及重大身体负荷的工伤日益受到关注，因为这些伤害会对工作效率、身体健康和工作幸福感产生负面影响，但也会因缺勤而产生经济成本。本观察性研究旨在量化码头工人使用可穿戴设备的人体工程学评估，以降低受伤风险。使用可穿戴动作捕捉系统分析和观察两种不同的手势:Task1A固定绑带和任务1B作为提升和锁定过程。还要求进行额外的测试和问卷调查:FMS、HHD、RAND-36、ODI和TSK问卷。结果发现，在不同的码头工人中，最高和最低学位以及他们工作的范围之间存在很大的差异。尽管存在很大的异质性，但我们发现在任务1B中，腕关节被保持在极端的位置，并且腰背的标准偏差与时间之间存在显著的相关性(R = 0.45, p = 0.045)。此外，所有参与者都表示有过抱怨;66%的人有运动恐惧症，60%的人超重，FMS的平均得分为15分。疲劳、抱怨、运动恐惧症、具有挑战性的工作环境或不健康的生活方式都可能导致运动模式的改变，这可能会增加受伤的风险。基于结果，我们建议实施热身和个性化的预防计划，以减少受伤的风险。
{"title":"Ergonomical assessment using wearable motion capture system in dockworkers during lashing","authors":"S. Scataglini, Lenie Denteneer, Nele Struyf, S. Truijen","doi":"10.1145/3597061.3597258","DOIUrl":"https://doi.org/10.1145/3597061.3597258","url":null,"abstract":"Work-related injuries involving significant physical loads are of growing concern, as these injuries can negatively impact work productivity, physical health and well-being at work, but also have economic costs due to absenteeism. This observational study aims to quantify the ergonomical assessment of dockworkers using wearable devices to reduce the risk of injuries. Two different gestures are analyzed and observed using a wearable mocap system: Task1A securing the lash bars and Task 1B as lifting and locking process. Additional tests and questionnaire were also requested: FMS, HHD, RAND-36, ODI, and TSK questionnaires. As result, it was found a large spread between the maximum and minimum degrees and the range in which they are working among the different dockworkers. Despite the large heterogeneity, we found for task 1B that the wrist was held in extreme positions, and a significant correlation was found between the time and standard deviation of the low back (R = 0.45, p = 0.045). In addition, all participants reported having complaints; 66% had kinesiophobia, 60% were overweight, and the mean score on the FMS was 15. The occurrence of fatigue, complaints, kinesiophobia, a challenging work environment, or an unhealthy lifestyle may lead to altered movement patterns, which could possibly increase the risk of injury. Based on the results, we recommend implementing a warm-up and an individualized prevention program to reduce the risk of injury.","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134277902","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}
Yusuke Sugimoto, Hamada Rizk, A. Uchiyama, Hirozumi Yamaguchi
Human Activity Recognition (HAR) has attracted considerable attention in recent years due to its potential applications in healthcare, smart homes, and security. Wi-Fi Channel State Information (CSI) is a promising sensor modality for HAR, providing a device-free and low-cost solution. However, building environment-independent models for HAR using Wi-Fi CSI remains a significant challenge. In this paper, we present a deep learning-based activity recognition system that exploits CSI measurements obtained from one or more environments to deliver consistent and accurate performance even in unseen environments. Our system employs a multi-task learning approach that is based on an encoder-decoder network architecture. This enables the encoder part of this architecture to mitigate the environment-dependent factors and extract a rich and environment-invariant representation. To evaluate the proposed system, we collected CSI samples for six activities pursued by three participants in four distinct environments. The results demonstrate the efficacy of the proposed system in achieving environment-independent HAR with an average accuracy of 80%. Additionally, the results validate the superiority of our method over environment-specific models by a minimum margin of 6% in cases of limited data.
{"title":"Towards Environment-Independent Activity Recognition Using Wi-Fi CSI with an Encoder-Decoder Network","authors":"Yusuke Sugimoto, Hamada Rizk, A. Uchiyama, Hirozumi Yamaguchi","doi":"10.1145/3597061.3597261","DOIUrl":"https://doi.org/10.1145/3597061.3597261","url":null,"abstract":"Human Activity Recognition (HAR) has attracted considerable attention in recent years due to its potential applications in healthcare, smart homes, and security. Wi-Fi Channel State Information (CSI) is a promising sensor modality for HAR, providing a device-free and low-cost solution. However, building environment-independent models for HAR using Wi-Fi CSI remains a significant challenge. In this paper, we present a deep learning-based activity recognition system that exploits CSI measurements obtained from one or more environments to deliver consistent and accurate performance even in unseen environments. Our system employs a multi-task learning approach that is based on an encoder-decoder network architecture. This enables the encoder part of this architecture to mitigate the environment-dependent factors and extract a rich and environment-invariant representation. To evaluate the proposed system, we collected CSI samples for six activities pursued by three participants in four distinct environments. The results demonstrate the efficacy of the proposed system in achieving environment-independent HAR with an average accuracy of 80%. Additionally, the results validate the superiority of our method over environment-specific models by a minimum margin of 6% in cases of limited data.","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134344328","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}
Sensor-equipped wearable devices are becoming increasingly popular in the healthcare industry, with some equipped with GPS and Proximity sensors as well. Raw (GPS) trajectories obtained through human-centric systems like body worn senors, and enriched with semantic annotations generate huge actionable insights for downstream domain specific applications like epidemic risk modeling. However, trajectory data suffer from missing data problem owing to various technical as well as behavioral factors. Our paper shows that, for a semantic trajectory dataset and using coarse grain semantic location for both prediction and imputation purposes, a simple ensemble classifier-based model can outperform the existing deep models where trajectory imputation is almost real-time delay.
{"title":"Imputation of Human Mobility Data for Comprehensive Risk Models","authors":"Shashee Kumari, Sakyajit Bhattacharya, Arnab Chatterjee, Avik Ghose","doi":"10.1145/3597061.3597260","DOIUrl":"https://doi.org/10.1145/3597061.3597260","url":null,"abstract":"Sensor-equipped wearable devices are becoming increasingly popular in the healthcare industry, with some equipped with GPS and Proximity sensors as well. Raw (GPS) trajectories obtained through human-centric systems like body worn senors, and enriched with semantic annotations generate huge actionable insights for downstream domain specific applications like epidemic risk modeling. However, trajectory data suffer from missing data problem owing to various technical as well as behavioral factors. Our paper shows that, for a semantic trajectory dataset and using coarse grain semantic location for both prediction and imputation purposes, a simple ensemble classifier-based model can outperform the existing deep models where trajectory imputation is almost real-time delay.","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130572281","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}
Smoking is a significant cause of death and deterioration of health worldwide, affecting active and passive smokers. Cessation of smoking contributes to an essential health and wellness application owing to the broad range of health problems such as cancer, hypertension, and several cardiopulmonary diseases. Personalized smoking-cessation applications can be very effective in helping users to stop smoking if there are detections and interventions done at the right time. This requires real-time detection of smoking puffs. Such applications are made feasible by day-long monitoring and smoking puff detection from unobtrusive devices such as wearables. This paper proposes a deep inference technique for the real-time detection of smoking puffs on a wearable device. We show that a simple, sequential Convolutional Neural Network (CNN) using only 6-axis Inertial signals can be utilized in place of complex and resource-consuming Deep Learning models. The accuracy achieved is comparable to State-of-the-Art techniques with an F1 score of 0.81, although the model size is tiny - 114 kB. Such small models can be deployed on the lowest configuration hardware platforms, achieving accurate but high-speed, low-power inference on conventional smartwatches. We ensure that the auto-designed models are directly compatible with resource-constrained platforms such as TensorFlow Lite and TensorFlow Lite for Microcontrollers (TFLM) without requiring further use of model reduction and optimization techniques. Our proposed approach will allow affordable wearable device manufacturers to run smoking detection on their devices, as it is tiny enough to fit TinyML platforms and is only dependent on IMU sensors that are universally available.
吸烟是世界范围内导致死亡和健康恶化的一个重要原因，对主动吸烟者和被动吸烟者都有影响。由于癌症、高血压和几种心肺疾病等广泛的健康问题，戒烟有助于基本的健康和保健应用。如果在适当的时候进行检测和干预，个性化的戒烟应用程序可以非常有效地帮助用户戒烟。这需要实时检测烟雾。这样的应用是可行的，全天监测和烟雾检测从不显眼的设备，如可穿戴设备。本文提出了一种用于可穿戴设备上烟雾实时检测的深度推理技术。我们展示了一个简单的、顺序的卷积神经网络(CNN)，它只使用6轴惯性信号，可以用来代替复杂的、消耗资源的深度学习模型。尽管模型大小很小(114 kB)，但所达到的精度与F1分数为0.81的最先进技术相当。这种小型模型可以部署在最低配置的硬件平台上，在传统智能手表上实现准确、高速、低功耗的推断。我们确保自动设计的模型直接兼容资源受限的平台，如TensorFlow Lite和TensorFlow Lite for微控制器(TFLM)，而不需要进一步使用模型缩减和优化技术。我们提出的方法将允许负担得起的可穿戴设备制造商在他们的设备上运行吸烟检测，因为它足够小，适合TinyML平台，并且只依赖于普遍可用的IMU传感器。
{"title":"TinyPuff: Automated design of Tiny Smoking Puff Classifiers for Body Worn Devices","authors":"Shalini Mukhopadhyay, Swarnava Dey, Avik Ghose","doi":"10.1145/3597061.3597259","DOIUrl":"https://doi.org/10.1145/3597061.3597259","url":null,"abstract":"Smoking is a significant cause of death and deterioration of health worldwide, affecting active and passive smokers. Cessation of smoking contributes to an essential health and wellness application owing to the broad range of health problems such as cancer, hypertension, and several cardiopulmonary diseases. Personalized smoking-cessation applications can be very effective in helping users to stop smoking if there are detections and interventions done at the right time. This requires real-time detection of smoking puffs. Such applications are made feasible by day-long monitoring and smoking puff detection from unobtrusive devices such as wearables. This paper proposes a deep inference technique for the real-time detection of smoking puffs on a wearable device. We show that a simple, sequential Convolutional Neural Network (CNN) using only 6-axis Inertial signals can be utilized in place of complex and resource-consuming Deep Learning models. The accuracy achieved is comparable to State-of-the-Art techniques with an F1 score of 0.81, although the model size is tiny - 114 kB. Such small models can be deployed on the lowest configuration hardware platforms, achieving accurate but high-speed, low-power inference on conventional smartwatches. We ensure that the auto-designed models are directly compatible with resource-constrained platforms such as TensorFlow Lite and TensorFlow Lite for Microcontrollers (TFLM) without requiring further use of model reduction and optimization techniques. Our proposed approach will allow affordable wearable device manufacturers to run smoking detection on their devices, as it is tiny enough to fit TinyML platforms and is only dependent on IMU sensors that are universally available.","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116845401","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}
D. Jaiswal, D. Chatterjee, Arindam Sarkar, M. S, R. K. Ramakrishnan, Arpan Pal, R. Ghosh
Differential dermal potentials is a non-invasive bio-potential signal acquired from the skin surface is a relatively new mode for human activity and behavior monitoring. Literature suggests acquiring differential dermal potentials from either hand or leg, which might be uncomfortable for long term usage. In the present work, for the first time we have acquired the differential dermal potentials from in and around the ear and have used that for monitoring mental workload. Experimental data were acquired from ear lobes and ear canals using two different devices while the participants were engaged in a mental arithmetic task. Thereafter, the signals are analyzed and a set of the most discriminating features are identified. Using these features, a cognitive load prediction model is developed. Results show that, for rest vs. load classification, we achieved an accuracy of 89% for ear canal data and 94.2% for ear lobe data. We also observed a positive co-occurrence of signals collected from these two locations. Thus, our proposed approach can be used for monitoring mental workload condition in real-life applications and is suitable for long term usage.
{"title":"Assessment of Mental Workloads using Differential Dermal Potentials Recorded from in and around Ear","authors":"D. Jaiswal, D. Chatterjee, Arindam Sarkar, M. S, R. K. Ramakrishnan, Arpan Pal, R. Ghosh","doi":"10.1145/3597061.3597257","DOIUrl":"https://doi.org/10.1145/3597061.3597257","url":null,"abstract":"Differential dermal potentials is a non-invasive bio-potential signal acquired from the skin surface is a relatively new mode for human activity and behavior monitoring. Literature suggests acquiring differential dermal potentials from either hand or leg, which might be uncomfortable for long term usage. In the present work, for the first time we have acquired the differential dermal potentials from in and around the ear and have used that for monitoring mental workload. Experimental data were acquired from ear lobes and ear canals using two different devices while the participants were engaged in a mental arithmetic task. Thereafter, the signals are analyzed and a set of the most discriminating features are identified. Using these features, a cognitive load prediction model is developed. Results show that, for rest vs. load classification, we achieved an accuracy of 89% for ear canal data and 94.2% for ear lobe data. We also observed a positive co-occurrence of signals collected from these two locations. Thus, our proposed approach can be used for monitoring mental workload condition in real-life applications and is suitable for long term usage.","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114072133","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}
{"title":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","authors":"","doi":"10.1145/3597061","DOIUrl":"https://doi.org/10.1145/3597061","url":null,"abstract":"","PeriodicalId":126710,"journal":{"name":"Proceedings of the 8th Workshop on Body-Centric Computing Systems","volume":"150 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124612929","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：