{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T06:03:51Z","timestamp":1784527431531,"version":"3.55.0"},"reference-count":49,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T00:00:00Z","timestamp":1784505600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Physiol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Accurately estimating physical activity (PA) and energy expenditure of children and adolescents cross-stage are key issues in sports medicine and beyond.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aims to develop and test a long short-term memory (LSTM) network to predict physical activity energy expenditure (PAEE), types, and intensity using accelerometer data collected in children and adolescents.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A total of 120 healthy participants aged 6\u201318 years completed 13 PA tasks while wearing triaxial accelerometers at 9 body sites and simultaneously wore the portable metabolic analyzer. 59 features were extracted from accelerometer signals, combined with demographic data to build LSTM models for cross-stage and single-stage (childhood, early and late adolescence), respectively, and compared with the results of 3 ensemble learning (random forest, extreme gradient boosting, and light gradient boosting machine).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The RMSE for predicting PAEE across and single stages were about 0.55 metabolic equivalents (METs) to 0.73 METs, and an average reduction of about 55% in RMSE over ensemble learning. The RMSE for predicting energy expenditure of a specific PA type was averaged about 0.38 METs to 0.49 METs, and an average reduction of about 60% in RMSE over ensemble learning. The recall for predicting PA types was about 90% to 93%, with an average relative improvement of about 70% over ensemble learning. The recall for predicted PA intensity was about 90% to 93%, with an average relative improvement of about 32% over ensemble learning.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>LSTM showed strong generalizability and robustness in cross-stage PA monitoring, offering an accurate and scalable solution.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fphys.2026.1852282","type":"journal-article","created":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T05:37:33Z","timestamp":1784525853000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Predicting physical activity energy expenditure, types, and intensity using long short-term memory networks in aged 6\u201318 cross-stage populations"],"prefix":"10.3389","volume":"17","author":[{"given":"Lei","family":"Jiang","sequence":"first","affiliation":[{"name":"Institute of Physical Education and Training, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanan","family":"Gao","sequence":"additional","affiliation":[{"name":"School of Physical Education and Sport Science, Fujian Normal University","place":["Fuzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunyu","family":"Zhao","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zice","family":"Guo","sequence":"additional","affiliation":[{"name":"Institute of Physical Education and Training, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhengyuan","family":"Huang","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qilin","family":"Fan","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junjun","family":"Lai","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rufeng","family":"Su","sequence":"additional","affiliation":[{"name":"Department of Exchange and Cooperation, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Youwei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinxi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Artificial Intelligence in Sports, Capital University of Physical Education and Sports","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,20]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1111\/joim.12908","article-title":"Measurement of physical activity in clinical practice using accelerometers","volume":"286","author":"Arvidsson","year":"2019","journal-title":"J. Intern. Med."},{"key":"B2","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1249\/MSS.0000000000001430","article-title":"A youth compendium of physical activities: Activity codes and metabolic intensities","volume":"50","author":"Butte","year":"2018","journal-title":"Med. Sci. Sports Exerc"},{"key":"B3","doi-asserted-by":"publisher","first-page":"3892","DOI":"10.1109\/TCYB.2016.2587723","article-title":"Weighted optimization-based distributed Kalman filter for nonlinear target tracking in collaborative sensor networks","volume":"47","author":"Chen","year":"2017","journal-title":"IEEE Trans. Cybern."},{"key":"B4","doi-asserted-by":"publisher","first-page":"1965","DOI":"10.1249\/MSS.0000000000001291","article-title":"Ensemble methods for classification of physical activities from wrist accelerometry","volume":"49","author":"Chowdhury","year":"2017","journal-title":"Med. Sci. Sports Exerc"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1994","DOI":"10.1249\/MSS.0b013e318219d939","article-title":"Identification of children\u2019s activity type with accelerometer-based neural networks","volume":"43","author":"De Vries","year":"2011","journal-title":"Med. Sci. Sports Exerc"},{"key":"B6","doi-asserted-by":"publisher","first-page":"3772","DOI":"10.1109\/TNNLS.2017.2741598","article-title":"Efficient online learning algorithms based on LSTM neural networks","volume":"29","author":"Ergen","year":"2018","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.2196\/18694","article-title":"Reliability and validity of commercially available wearable devices for measuring steps, energy expenditure, and heart rate: Systematic review","volume":"8","author":"Fuller","year":"2020","journal-title":"JMIR Mhealth Uhealth"},{"key":"B8","doi-asserted-by":"publisher","first-page":"329","DOI":"10.1249\/01.mss.0000153115.33762.3f","article-title":"Energy costs of physical activities in children and adolescents","volume":"37","author":"Harrell","year":"2005","journal-title":"Med. Sci. Sports Exerc"},{"key":"B9","doi-asserted-by":"publisher","first-page":"2791","DOI":"10.1109\/TNNLS.2018.2890117","article-title":"Fast matrix factorization with nonuniform weights on missing data","volume":"31","author":"He","year":"2020","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B10","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1109\/JBHI.2020.2982935","article-title":"An adaptive Kalman filter bank for ECG denoising","volume":"25","author":"Hesar","year":"2021","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"B11","doi-asserted-by":"publisher","first-page":"846","DOI":"10.1249\/MSS.0000000000001486","article-title":"Adapted sojourn models to estimate activity intensity in youth: A suite of tools","volume":"50","author":"Hibbing","year":"2018","journal-title":"Med. Sci. Sports Exerc"},{"key":"B12","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1186\/s12966-023-01515-0","article-title":"Criterion validity of wrist accelerometry for assessing energy intake via the intake-balance technique","volume":"20","author":"Hibbing","year":"2023","journal-title":"Int. J. Behav. Nutr. Phys. Act."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1816","DOI":"10.1249\/MSS.0000000000000289","article-title":"Age group comparability of raw accelerometer output from wrist- and hip-worn monitors","volume":"46","author":"Hildebrand","year":"2014","journal-title":"Med. Sci. Sports Exerc"},{"key":"B14","doi-asserted-by":"publisher","first-page":"2345","DOI":"10.1109\/JBHI.2023.3252173","article-title":"Energy expenditure estimation from respiratory magnetometer plethysmography: A comparison study","volume":"27","author":"Houssein","year":"2023","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.2196\/30135","article-title":"Automatic recognition and analysis of balance activity in community-dwelling older adults: Algorithm validation","volume":"23","author":"Hsu","year":"2021","journal-title":"J. Med. Internet Res."},{"key":"B16","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1186\/s12984-020-00798-9","article-title":"Inertial sensor-based gait parameters reflect patient-reported fatigue in multiple sclerosis","volume":"17","author":"Ibrahim","year":"2020","journal-title":"J. NeuroEng. Rehabil."},{"key":"B17","doi-asserted-by":"publisher","first-page":"1776","DOI":"10.1249\/MSS.0b013e3181d479f2","article-title":"Accelerometer output and MET values of common physical activities","volume":"42","author":"Kozey","year":"2010","journal-title":"Med. Sci. Sports Exerc"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2086","DOI":"10.1109\/JBHI.2021.3118717","article-title":"Transfer learning improves accelerometer-based child activity recognition via subject-independent adult-domain adaption","volume":"26","author":"Li","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.2196\/11201","article-title":"Applying multivariate segmentation methods to human activity recognition from wearable sensors\u2019 data","volume":"7","author":"Li","year":"2019","journal-title":"JMIR Mhealth Uhealth"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.2196\/59497","article-title":"Data analytics in physical activity studies\u00a0with accelerometers: Scoping review","volume":"26","author":"Liang","year":"2024","journal-title":"J. Med. Internet Res."},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1016\/j.jshs.2024.100973","article-title":"Accelerometer-measured physical activity, sedentary behavior, and incidence of macrovascular and microvascular events in individuals with type 2 diabetes mellitus and prediabetes","volume":"14","author":"Liang","year":"2024","journal-title":"J.\u00a0Sport Health Sci."},{"key":"B22","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1038\/s41572-023-00435-4","article-title":"Child and adolescent obesity","volume":"9","author":"Lister","year":"2023","journal-title":"Nat. Rev. Dis. Primers"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1821","DOI":"10.1007\/s40279-017-0716-0","article-title":"Accelerometer data collection and processing criteria to assess physical activity and other outcomes: A systematic review and practical considerations","volume":"47","author":"Migueles","year":"2017","journal-title":"Sports Med."},{"key":"B24","doi-asserted-by":"publisher","first-page":"590","DOI":"10.1249\/MSS.0000000000001797","article-title":"Accelerometer data processing and energy expenditure estimation in preschoolers","volume":"51","author":"Migueles","year":"2019","journal-title":"Med. Sci. Sports Exerc"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1249\/MSS.0000000000000597","article-title":"Energy expenditure prediction using raw accelerometer data in simulated free living","volume":"47","author":"Montoye","year":"2015","journal-title":"Med. Sci. Sports Exerc"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.2196\/23938","article-title":"Comparison of the validity and generalizability of machine learning algorithms for the prediction of energy expenditure: Validation study","volume":"9","author":"O\u2019Driscoll","year":"2021","journal-title":"JMIR Mhealth Uhealth"},{"key":"B27","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1016\/j.jshs.2020.02.004","article-title":"Activity-related typologies and longitudinal change in physical activity and sedentary time in children and adolescents: The UP&DOWN study","volume":"10","author":"Parker","year":"2021","journal-title":"J. Sport Health Sci."},{"key":"B28","doi-asserted-by":"publisher","first-page":"1725","DOI":"10.1161\/CIRCRESAHA.123.322121","article-title":"Physical activity over the lifecourse and cardiovascular disease","volume":"132","author":"Perry","year":"2023","journal-title":"Circ. Res."},{"key":"B29","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1186\/s12984-024-01327-8","article-title":"Emerging methods for measuring physical activity using accelerometry in children and adolescents with neuromotor disorders: A narrative review","volume":"21","author":"Petersen","year":"2024","journal-title":"J. NeuroEng. Rehabil."},{"key":"B30","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1186\/s12966-020-01067-7","article-title":"Resemblance in accelerometer-assessed physical activity in families with children: The Lolland-Falster Health Study","volume":"17","author":"Petersen","year":"2020","journal-title":"Int. J. Behav. Nutr. Phys. Act."},{"key":"B31","doi-asserted-by":"publisher","first-page":"808","DOI":"10.1126\/science.abe5017","article-title":"Daily energy expenditure through the human life course","volume":"373","author":"Pontzer","year":"2021","journal-title":"Science"},{"key":"B32","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1146\/annurev-publhealth-031210-101151","article-title":"Physical activity for health: What\u00a0kind? How much? How intense? On top of what","volume":"32","author":"Powell","year":"2011","journal-title":"Annu. Rev. Public Health"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1016\/j.redox.2024.103470","article-title":"Machine learning and SHAP value interpretation for predicting comorbidity of cardiovascular disease and cancer with dietary antioxidants","volume":"79","author":"Qi","year":"2025","journal-title":"Redox Biol."},{"key":"B34","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1249\/MSS.0000000000000778","article-title":"24 hours of sleep, sedentary behavior, and physical activity with nine wearable devices","volume":"48","author":"Rosenberger","year":"2016","journal-title":"Med. Sci. Sports Exerc"},{"key":"B35","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1249\/MSS.0000000000000394","article-title":"Comparability of measured acceleration from accelerometry-based activity monitors","volume":"47","author":"Rowlands","year":"2015","journal-title":"Med. Sci. Sports Exerc"},{"key":"B36","doi-asserted-by":"publisher","first-page":"1452","DOI":"10.1109\/TNNLS.2018.2869822","article-title":"Nonuniformly sampled data processing using LSTM networks","volume":"30","author":"Sahin","year":"2019","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.2196\/mhealth.6562","article-title":"Sleep quality prediction from wearable data using deep learning","volume":"4","author":"Sathyanarayana","year":"2016","journal-title":"JMIR Mhealth Uhealth"},{"key":"B38","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1186\/s12984-021-00958-5","article-title":"Prediction and detection of freezing of gait in Parkinson\u2019s disease from plantar pressure data using long short-term memory neural-networks","volume":"18","author":"Shalin","year":"2021","journal-title":"J. NeuroEng. Rehabil."},{"key":"B39","doi-asserted-by":"publisher","first-page":"439","DOI":"10.2165\/00007256-200131060-00004","article-title":"Physical activity assessment in children and adolescents","volume":"31","author":"Sirard","year":"2001","journal-title":"Sports Med."},{"key":"B40","doi-asserted-by":"publisher","first-page":"4312","DOI":"10.1038\/s41467-021-24173-x","article-title":"Sensing leg movement enhances wearable monitoring of energy expenditure","volume":"12","author":"Slade","year":"2021","journal-title":"Nat. Commun."},{"key":"B41","doi-asserted-by":"publisher","first-page":"2477","DOI":"10.1007\/s40279-023-01905-1","article-title":"The role of multi-sensor measurement in the assessment of movement quality: A systematic review","volume":"53","author":"Swain","year":"2023","journal-title":"Sports Med."},{"key":"B42","doi-asserted-by":"publisher","first-page":"1801","DOI":"10.1249\/MSS.0b013e318258ac11","article-title":"Artificial neural networks to predict activity type and energy expenditure in youth","volume":"44","author":"Trost","year":"2012","journal-title":"Med. Sci. Sports Exerc"},{"key":"B43","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1186\/s12966-018-0651-y","article-title":"Cadence (steps\/min) and intensity during ambulation in 6-20 year olds: The CADENCE-kids study","volume":"15","author":"Tudor-Locke","year":"2018","journal-title":"Int. J. Behav. Nutr. Phys. Act."},{"key":"B44","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1016\/j.gaitpost.2020.05.010","article-title":"Normative database of spatiotemporal gait parameters using inertial sensors in typically developing children and young adults","volume":"80","author":"Voss","year":"2020","journal-title":"Gait Posture"},{"key":"B45","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1016\/j.isatra.2021.12.041","article-title":"Constrained two-stage Kalman filter for real-time state estimation of systems with time-varying measurement noise covariance","volume":"129","author":"Wang","year":"2022","journal-title":"ISA Trans."},{"key":"B46","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1249\/MSS.0000000000003294","article-title":"A device agnostic approach to predict children\u2019s activity from consumer wearable accelerometer data: A proof-of-concept study","volume":"56","author":"Weaver","year":"2024","journal-title":"Med. Sci. Sports Exerc"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.3390\/nu15224821","article-title":"Associations between high protein intake, linear growth, and stunting in children and adolescents: A cross-sectional study","volume":"15","author":"Xiong","year":"2023","journal-title":"Nutrients"},{"key":"B48","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1152\/japplphysiol.00729.2009","article-title":"Multivariate adaptive regression splines models for the prediction of energy expenditure in children and adolescents","volume":"108","author":"Zakeri","year":"2010","journal-title":"J. Appl. Physiol. (1985)"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2023.106157","article-title":"Improved LSTM-based deep learning model for COVID-19 prediction using optimized approach","volume":"122","author":"Zhou","year":"2023","journal-title":"Eng. Appl. Artif. Intell."}],"container-title":["Frontiers in Physiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphys.2026.1852282\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T05:37:34Z","timestamp":1784525854000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fphys.2026.1852282\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,20]]},"references-count":49,"alternative-id":["10.3389\/fphys.2026.1852282"],"URL":"https:\/\/doi.org\/10.3389\/fphys.2026.1852282","relation":{},"ISSN":["1664-042X"],"issn-type":[{"value":"1664-042X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,20]]},"article-number":"1852282"}}