[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124854-en":3,"doc-seo-124854-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124854,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Annotating Sleep States in Children from Wrist-Worn Accelerometer Data Using Machine Learning","Sleep detection and annotation are essential for researchers studying sleep patterns, particularly in children. Modern wrist-worn watches with built-in accelerometers enable collection of sleep-related logs, but converting them into accurate sleep events such as sleep onset and wakeup remains difficult. The work proposes automated, scalable labeling using multiple machine learning models, from support vectors and boosting to ensemble methods and sequence models like LSTMs and Region-based CNNs. Model quality is evaluated using Event Detection Average Precision (EDAP).","Annotating sleep states in children from wrist-worn accelerometer data using  \nMachine Learning  \nAshwin Ram * 1 Sundar Sripada V. S. * 1 Shuvam Keshari * 1 Zizhe Jiang 2  \narXiv :2312 .07561v1 [ ee ss . SP] 9 Dec 2023  \nAbstract  \nSleep detection and annotation are crucial for researchers to understand sleep patterns, especially in children. With modern wrist-worn watches comprising built-in accelerometers, sleep logs can be collected. However, the annotation of these logs into distinct sleep events – onset and wakeup, proves to be challenging. These annotations must be automated, precise, and scalable. We propose to model the accelerometer data using different machine learning (ML) techniques such as support vectors, boosting, ensemble methods, and more complex approaches involving LSTMs and Region-based CNNs. Later, we aim to evaluate these approaches using the Event Detection Average Precision (EDAP) score (similar to the IOU metric) to eventually compare predictive power and model performance. Our GitHub repository can be viewed here.  \n1. Introduction  \n1.1. Why is sleep detection important?  \nSleep is crucial in regulating mood, emotions, and behavior in individuals of all ages, particularly children [1] . By accurately detecting periods of sleep and wakefulness from wrist-worn accelerometer data, researchers can gain a deeper understanding of sleep patterns and better understand disturbances caused by sleep in children [2, 3] .  \n1.2. How does data science play a role in sleep detection?  \nSo far in sleep literature, the most accurate way to annotate sleep events is by using sleep logs. However, the manual effort required to maintain these logs over time engenders nuances in the annotation; a person going to bed versus  \n*Equal contribution 1 Chandra Department of Electrical and Computer Engineering, UT Austin 2Department of Operations Research and Industrial Engineering, UT Austin. Correspondence to: Shuvam Keshari \u003C[skeshari@utexas.edu](skeshari@utexas.edu) > .  \nCopyright 2023 by the author(s) .  \nactually falling asleep may be at very different times, which may fabricate errors in these logs [2] .  \nIn contrast, a wrist-worn accelerometer can track humanengineered sleep-related features (like arm angle) which can potentially better help detect and annotate sleep events. However, these features vary across individuals making it hard for sleep researchers to accurately identify sleep windows. By leveraging the right Machine Learning (ML) approaches, it is possible to learn the higher dimensional sleep data and accurately annotate sleep events, thereby bridging the gap between data availability and data analyzability [4, 5] .  \n1.3. Related Work  \nThe rapidly evolving domain of sleep pattern detection through wearable technology has gained considerable traction in recent times. A standout contribution in this area is by [6], who have ingeniously employed smartphones and wearables to introduce a non-intrusive method for ambulatory sleep tracking. Leveraging the power of Long Short-Term Memory (LSTM) Recurrent Neural Networks, they have adeptly determined sleep/wake states by assimilating temporal information. Their exhaustive research, spanning 5580 days and involving 186 participants, reported an admirable sleep/wake classification accuracy of 96 .5% . Remarkably, when pitted against traditional non-temporal machine learning techniques and well-regarded actigraphy software, the LSTM model demonstrated unparalleled efficacy, signifying the substantial potential of recurrent neural networks in sleep detection. Reflecting on the ongoing Kaggle competition, many teams are extensively engineering features on ‘train’ and ‘test’ datasets, originally extracted from Parquet files. They craft temporal attributes from the’timestamp’ column, optimize memory usage, and meticulously introduce new features employing statistical techniques, including rolling metrics and lag/lead shifts. This intricate feature engineering is orchestra","cbCaineBuGvWXUd8","https://ap.wps.com/l/cbCaineBuGvWXUd8","pdf",2147608,1,7,"English","en",105,"# Abstract\n# Introduction\n## Why is sleep detection important?\n## How does data science play a role in sleep detection?\n## Related Work\n# Problem Description\n## Data Explanation","[{\"question\":\"Why is sleep detection important for children?\",\"answer\":\"Accurate detection of sleep and wakefulness from wrist-worn accelerometer data helps researchers understand children’s sleep patterns and the disturbances sleep can cause.\"},{\"question\":\"What makes sleep log annotation difficult in wearable data?\",\"answer\":\"Manual sleep logs introduce timing-related inconsistencies, and distinguishing sleep windows can be challenging because sleep-related features vary across individuals.\"},{\"question\":\"Which machine learning approaches are proposed for sleep event annotation?\",\"answer\":\"The approach models accelerometer data using techniques such as support vector methods, boosting, ensemble methods, and more complex models including LSTMs and Region-based CNNs.\"}]","Annotating Sleep States in Children from Wrist-Worn Accelerometer Data Using Machine Learning | PDF",1785895043,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"annotating-sleep-states-in-children-from-wrist-worn-accelerometer-data-using-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/annotating-sleep-states-in-children-from-wrist-worn-accelerometer-data-using-machine-learning/124854/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is sleep detection important for children?","Question",{"text":75,"@type":76},"Accurate detection of sleep and wakefulness from wrist-worn accelerometer data helps researchers understand children’s sleep patterns and the disturbances sleep can cause.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What makes sleep log annotation difficult in wearable data?",{"text":80,"@type":76},"Manual sleep logs introduce timing-related inconsistencies, and distinguishing sleep windows can be challenging because sleep-related features vary across individuals.",{"name":82,"@type":73,"acceptedAnswer":83},"Which machine learning approaches are proposed for sleep event annotation?",{"text":84,"@type":76},"The approach models accelerometer data using techniques such as support vector methods, boosting, ensemble methods, and more complex models including LSTMs and Region-based CNNs.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]