{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T16:04:18Z","timestamp":1789142658624,"version":"build-2803163510"},"reference-count":28,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2024,3,7]],"date-time":"2024-03-07T00:00:00Z","timestamp":1709769600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Direct Grant for Research, The Chinese University of Hong Kong"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:sec id=\"sec001\">\n<jats:title>Background<\/jats:title>\n<jats:p>Accelerometers are widely adopted in research and consumer devices as a tool to measure physical activity. However, existing algorithms used to estimate activity intensity are wear-site-specific. Non-compliance to wear instructions may lead to misspecifications. In this study, we developed deep neural network models to classify device placement and activity intensity based on raw acceleration data. Performances of these models were evaluated by making comparisons to the ground truth and results derived from existing count-based algorithms.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec002\">\n<jats:title>Methods<\/jats:title>\n<jats:p>54 participants (26 adults 26.9\u00b18.7 years; 28 children, 12.1\u00b12.3 years) completed a series of activity tasks in a laboratory with accelerometers attached to each of their hip, wrist, and chest. Their metabolic rates at rest and during activity periods were measured using the portable COSMED K5; data were then converted to metabolic equivalents, and used as the ground truth for activity intensity. Deep neutral networks using the Long Short-Term Memory approach were trained and evaluated based on raw acceleration data collected from accelerometers. Models to classify wear-site and activity intensity, respectively, were evaluated.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec003\">\n<jats:title>Results<\/jats:title>\n<jats:p>The trained models correctly classified wear-sites and activity intensities over 90% of the time, which outperformed count-based algorithms (wear-site correctly specified: 83% to 85%; wear-site misspecified: 64% to 75%). When additional parameters of age, height and weight of participants were specified, the accuracy of some prediction models surpassed 95%.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec004\">\n<jats:title>Conclusions<\/jats:title>\n<jats:p>Results of the study suggest that accelerometer placement could be determined prospectively, and non-wear-site-specific algorithms had satisfactory accuracies. The performances, in terms of intensity classification, of these models also exceeded typical count-based algorithms. Without being restricted to one specific wear-site, research protocols for accelerometers wear could allow more autonomy to participants, which may in turn improve their acceptance and compliance to wear protocols, and in turn more accurate results.<\/jats:p>\n<\/jats:sec>","DOI":"10.1371\/journal.pone.0299295","type":"journal-article","created":{"date-parts":[[2024,3,7]],"date-time":"2024-03-07T18:54:31Z","timestamp":1709837671000},"page":"e0299295","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":15,"title":["Development of a multi-wear-site, deep learning-based physical activity intensity classification algorithm using raw acceleration data"],"prefix":"10.1371","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0783-879X","authenticated-orcid":true,"given":"Johan Y. Y.","family":"Ng","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joni H.","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stanley S.","family":"Hui","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guanxian","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fung","family":"Yau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9943-9122","authenticated-orcid":true,"given":"Amy S.","family":"Ha","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2024,3,7]]},"reference":[{"key":"pone.0299295.ref001","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1186\/1479-5868-7-40","article-title":"Systematic review of the health benefits of physical activity and fitness in school-aged children and youth.","volume":"7","author":"I Janssen","year":"2010","journal-title":"Int J Behav Nutr Phys Act"},{"key":"pone.0299295.ref002","doi-asserted-by":"crossref","first-page":"146","DOI":"10.1016\/j.psychsport.2018.08.011","article-title":"Physical activity and mental health in children and adolescents: An updated review of reviews and an analysis of causality.","volume":"42","author":"SJH Biddle","year":"2019","journal-title":"Psychol Sport Exerc"},{"issue":"5","key":"pone.0299295.ref003","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1097\/HCO.0000000000000437","article-title":"Health benefits of physical activity: a systematic review of current systematic reviews.","volume":"32","author":"DER Warburton","year":"2017","journal-title":"Curr Opin Cardiol"},{"key":"pone.0299295.ref004","doi-asserted-by":"crossref","first-page":"106541","DOI":"10.1016\/j.cmpb.2021.106541","article-title":"Application of artificial intelligence in wearable devices: Opportunities and challenges","volume":"213","author":"D Nahavandi","year":"2022","journal-title":"Comput Meth Prog Bio"},{"key":"pone.0299295.ref005","doi-asserted-by":"crossref","first-page":"e110","DOI":"10.2196\/jmir.9157","article-title":"Using fitness trackers and smartwatches to measure physical activity in research: Analysis of consumer wrist-worn wearables","volume":"20","author":"A Henriksen","year":"2018","journal-title":"J Med Internet Res"},{"key":"pone.0299295.ref006","doi-asserted-by":"crossref","first-page":"e11819","DOI":"10.2196\/11819","article-title":"Consumer-based wearable activity trackers increase physical activity participation: Systematic review and meta-analysis.","volume":"7","author":"KJ Brickwood","year":"2019","journal-title":"JMIR mHealth uHealth."},{"issue":"8","key":"pone.0299295.ref007","doi-asserted-by":"crossref","first-page":"E615","DOI":"10.1016\/S2589-7500(22)00111-X","article-title":"Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses","volume":"4","author":"T Ferguson","year":"2022","journal-title":"Lancet Digit Health"},{"issue":"14","key":"pone.0299295.ref008","doi-asserted-by":"crossref","first-page":"1557","DOI":"10.1080\/02640410802334196","article-title":"Calibration of two objective measures of physical activity for children.","volume":"26","author":"KR Evenson","year":"2006","journal-title":"J Sports Sci."},{"issue":"5","key":"pone.0299295.ref009","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1097\/00005768-199805000-00021","article-title":"Calibration of the Computer Science and Applications, Inc. accelerometer","volume":"30","author":"PS Freedson","year":"1998","journal-title":"Med Sci Sports Exerc."},{"issue":"9","key":"pone.0299295.ref010","doi-asserted-by":"crossref","first-page":"1036","DOI":"10.1080\/02640414.2020.1740469","article-title":"Development of raw acceleration cut-points for wrist and hip accelerometers to assess sedentary behaviour and physical activity in 5\u20137-year-old children.","volume":"38","author":"M Crotti","year":"2020","journal-title":"J Sports Sci."},{"issue":"8","key":"pone.0299295.ref011","doi-asserted-by":"crossref","first-page":"e0160644","DOI":"10.1371\/journal.pone.0160644","article-title":"An activity index for raw accelerometry data and its comparison with other activity metrics.","volume":"11","author":"J Bai","year":"2016","journal-title":"PLoS ONE"},{"key":"pone.0299295.ref012","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1016\/j.gaitpost.2017.12.028","article-title":"Calibration of raw accelerometer data to measure physical activity: A systematic review.","volume":"61","author":"M de Almeida Mendes","year":"2018","journal-title":"Gait & Posture."},{"key":"pone.0299295.ref013","doi-asserted-by":"crossref","first-page":"285","DOI":"10.1016\/j.gaitpost.2018.12.003","article-title":"Calibration and validation of accelerometer-based activity monitors: A systematic review of machine-learning approaches.","volume":"68","author":"V Farrahi","year":"2019","journal-title":"Gait & Posture."},{"issue":"2","key":"pone.0299295.ref014","doi-asserted-by":"crossref","first-page":"298","DOI":"10.3934\/publichealth.2016.2.298","article-title":"Validation and comparison of accelerometers worn on the hip, thigh, and wrists for measuring physical activity and sedentary behavior.","volume":"3","author":"AHK Montoye","year":"2016","journal-title":"AIMS Public Health"},{"issue":"3","key":"pone.0299295.ref015","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1123\/jpah.2019-0088","article-title":"Application of raw accelerometer data and machine-learning techniques to characterize human movement behavior: A systematic scoping review","volume":"17","author":"A Narayanan","year":"2020","journal-title":"J Phys Act Health"},{"issue":"1","key":"pone.0299295.ref016","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1123\/jmpb.2021-0015","article-title":"A machine learning classifier for detection of physical activity types and postures during free-living.","volume":"5","author":"K Bach","year":"2021","journal-title":"J Meas Phys Behav"},{"key":"pone.0299295.ref017","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.medengphy.2017.12.006","article-title":"Automatic recognition of gait patterns in human motor disorders using machine learning: A review.","volume":"53","author":"J Figueiredo","year":"2018","journal-title":"Med Eng Phys"},{"key":"pone.0299295.ref018","doi-asserted-by":"crossref","first-page":"951","DOI":"10.1111\/coin.12441","article-title":"A systematic review on machine learning for fall detection system.","volume":"37","author":"S Rastogi","year":"2021","journal-title":"Comput Intell"},{"key":"pone.0299295.ref019","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1016\/j.gaitpost.2015.11.005","article-title":"Comparison of physical activity assessed using hip- and wrist-worn accelerometers.","volume":"44","author":"M Kamada","year":"2016","journal-title":"Gait & Posture."},{"issue":"4","key":"pone.0299295.ref020","doi-asserted-by":"crossref","first-page":"368","DOI":"10.1016\/j.jsams.2016.04.013","article-title":"Accelerometer wear-site detection: When one site does not suit all, all of the time.","volume":"20","author":"AV Rowlands","year":"2017","journal-title":"J Sci Med Sport."},{"issue":"12","key":"pone.0299295.ref021","doi-asserted-by":"crossref","first-page":"1101","DOI":"10.1016\/j.jsams.2017.04.017","article-title":"Comparability and feasibility of wrist-and hip-worn accelerometers in free-living adolescents.","volume":"20","author":"JJ Scott","year":"2017","journal-title":"J Sci Med Sport"},{"issue":"12","key":"pone.0299295.ref022","doi-asserted-by":"crossref","first-page":"1015","DOI":"10.1016\/j.jsams.2016.03.004","article-title":"Feasibility of a chest-worn accelerometer for physical activity measurement.","volume":"19","author":"JH Zhang","year":"2016","journal-title":"J Sci Med Sport"},{"key":"pone.0299295.ref023","doi-asserted-by":"crossref","first-page":"27","DOI":"10.3390\/informatics5020027","article-title":"A comprehensive study of activity recognition using accelerometers.","volume":"5","author":"N Twomey","year":"2018","journal-title":"Informatics"},{"issue":"7","key":"pone.0299295.ref024","doi-asserted-by":"crossref","first-page":"1644","DOI":"10.3390\/s19071644","article-title":"Accelerometer-based human fall detection using convolutional neural networks.","volume":"19","author":"GL Santos","year":"2019","journal-title":"Sensors"},{"key":"pone.0299295.ref025","doi-asserted-by":"crossref","first-page":"e000874","DOI":"10.1136\/bmjsem-2020-000874","article-title":"Thigh-worn accelerometry for measuring movement and posture across the 24-hour cycle: a scoping review and expert statement","volume":"6","author":"ML Stevens","year":"2020","journal-title":"BMJ Open Sport Exerc Med"},{"key":"pone.0299295.ref026","doi-asserted-by":"crossref","first-page":"3154","DOI":"10.3390\/s21093154","article-title":"Validity of Hip and Ankle Worn Actigraph Accelerometers for Measuring Steps as a Function of Gait Speed during Steady State Walking and Continuous Turning.","volume":"21","author":"L Bezuidenhout","year":"2021","journal-title":"Sensors"},{"key":"pone.0299295.ref027","doi-asserted-by":"crossref","first-page":"724714","DOI":"10.3389\/fdgth.2021.724714","article-title":"Activity Tracking Using Ear-Level Accelerometers.","volume":"3","author":"MA Skoglund","year":"2021","journal-title":"Front Digit Health."},{"key":"pone.0299295.ref028","doi-asserted-by":"crossref","first-page":"253","DOI":"10.1186\/s12890-019-0942-3","article-title":"A web-based intervention to promote physical activity in adolescents and young adults with cystic fibrosis: protocol for a randomized controlled trial.","volume":"19","author":"NS Cox","year":"2019","journal-title":"BMC Pulm Med"}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0299295","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,3,7]],"date-time":"2024-03-07T18:54:47Z","timestamp":1709837687000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0299295"}},"subtitle":[],"editor":[{"given":"Yaodong","family":"Gu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,3,7]]},"references-count":28,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2024,3,7]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0299295","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,3,7]]}}}