{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,9]],"date-time":"2026-01-09T00:49:00Z","timestamp":1767919740038,"version":"3.49.0"},"reference-count":50,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2022,4,15]],"date-time":"2022-04-15T00:00:00Z","timestamp":1649980800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Institute of Health","award":["R01AG042525"],"award-info":[{"award-number":["R01AG042525"]}]},{"name":"Claude D. Pepper Older Americans Independence Centers at the University of Florida","award":["P30AG028740"],"award-info":[{"award-number":["P30AG028740"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Sufficient physical activity (PA) reduces the risk of a myriad of diseases and preserves physical capabilities in later life. While there have been significant achievements in mapping accelerations to real-life movements using machine learning (ML), errors continue to be common, particularly for wrist-worn devices. It remains unknown whether ML models are robust for estimating age-related loss of physical function. In this study, we evaluated the performance of ML models (XGBoost and LASSO) to estimate the hallmark measures of PA in low physical performance (LPP) and high physical performance (HPP) groups. Our models were built to recognize PA types and intensities, identify each individual activity, and estimate energy expenditure (EE) using wrist-worn accelerometer data (33 activities per participant) from a large sample of participants (n = 247, 57% females, aged 60+ years). Results indicated that the ML models were accurate in recognizing PA by type and intensity while also estimating EE accurately. However, the models built to recognize individual activities were less robust. Across all tasks, XGBoost outperformed LASSO. XGBoost obtained F1-Scores for sedentary (0.932 \u00b1 0.005), locomotion (0.946 \u00b1 0.003), lifestyle (0.927 \u00b1 0.006), and strength flexibility exercise (0.915 \u00b1 0.017) activity type recognition tasks. The F1-Scores for recognizing low, light, and moderate activity intensity were (0.932\u00a0\u00b1\u00a00.005), (0.840 \u00b1 0.004), and (0.869 \u00b1 0.005), respectively. The root mean square error for EE estimation was 0.836 \u00b1 0.059 METs. There was no evidence showing that splitting the participants into the LPP and HPP groups improved the models\u2019 performance on estimating the hallmark measures of physical activities. In conclusion, using features derived from wrist-worn accelerometer data, machine learning models can accurately recognize PA types and intensities and estimate EE for older adults with high and low physical function.<\/jats:p>","DOI":"10.3390\/s22083061","type":"journal-article","created":{"date-parts":[[2022,4,19]],"date-time":"2022-04-19T02:39:31Z","timestamp":1650335971000},"page":"3061","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Are Machine Learning Models on Wrist Accelerometry Robust against Differences in Physical Performance among Older Adults?"],"prefix":"10.3390","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0961-1927","authenticated-orcid":false,"given":"Chen","family":"Bai","sequence":"first","affiliation":[{"name":"Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, FL 32610, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5763-5184","authenticated-orcid":false,"given":"Amal A.","family":"Wanigatunga","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD 21205, USA"}]},{"given":"Santiago","family":"Saldana","sequence":"additional","affiliation":[{"name":"Department of Biostatistics and Data Science, School of Medicine, Wake Forest University, Winston-Salem, NC 27101, USA"}]},{"given":"Ramon","family":"Casanova","sequence":"additional","affiliation":[{"name":"Department of Biostatistics and Data Science, School of Medicine, Wake Forest University, Winston-Salem, NC 27101, USA"}]},{"given":"Todd M.","family":"Manini","sequence":"additional","affiliation":[{"name":"Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, FL 32610, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5345-8811","authenticated-orcid":false,"given":"Mamoun T.","family":"Mardini","sequence":"additional","affiliation":[{"name":"Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida, Gainesville, FL 32610, USA"}]}],"member":"1968","published-online":{"date-parts":[[2022,4,15]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"801","DOI":"10.1249\/MSS.0000000000001144","article-title":"Activity recognition in youth using single accelerometer placed at wrist or ankle","volume":"49","author":"Mannini","year":"2017","journal-title":"Med. Sci. Sports Exerc."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Chernbumroong, S., Atkins, A.S., and Yu, H. (2011, January 8\u201311). Activity classification using a single wrist-worn accelerometer. Proceedings of the 2011 5th International Conference on Software, Knowledge Information, Industrial Management and Applications (SKIMA) Proceedings, Benevento, Italy.","DOI":"10.1109\/SKIMA.2011.6089975"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Weiss, G.M., Timko, J.L., Gallagher, C.M., Yoneda, K., and Schreiber, A.J. (2016, January 24\u201327). Smartwatch-based activity recognition: A machine learning approach. Proceedings of the 2016 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), Las Vegas, NV, USA.","DOI":"10.1109\/BHI.2016.7455925"},{"key":"ref_4","unstructured":"Gjoreski, H., Bizjak, J., Gjoreski, M., and Gams, M. (2016, January 9). Comparing deep and classical machine learning methods for human activity recognition using wrist accelerometer. Proceedings of the IJCAI 2016 Workshop on Deep Learning for Artificial Intelligence, New York, NY, USA."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Panwar, M., Ram Dyuthi, S., Chandra Prakash, K., Biswas, D., Acharyya, A., Maharatna, K., Gautam, A., and Naik, G.R. (2017, January 11\u201315). CNN based approach for activity recognition using a wrist-worn accelerometer. Proceedings of the 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Jeju, Korea.","DOI":"10.1109\/EMBC.2017.8037349"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Kheirkhahan, M., Mehta, S., Nath, M., Wanigatunga, A.A., Corbett, D.B., Manini, T.M., and Ranka, S. (2017, January 13\u201316). A bag-of-words approach for assessing activities of daily living using wrist accelerometer data. Proceedings of the 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Kansas City, MO, USA.","DOI":"10.1109\/BIBM.2017.8217735"},{"key":"ref_7","doi-asserted-by":"crossref","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":"ref_8","doi-asserted-by":"crossref","unstructured":"Dijkhuis, T.B., Blaauw, F.J., van Ittersum, M.W., Velthuijsen, H., and Aiello, M. (2018). Personalized physical activity coaching: A machine learning approach. Sensors, 18.","DOI":"10.3390\/s18020623"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Ka\u0144toch, E. (2018). Recognition of sedentary behavior by machine learning analysis of wearable sensors during activities of daily living for telemedical assessment of cardiovascular risk. Sensors, 18.","DOI":"10.3390\/s18103219"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Chowdhury, A.K., Tjondronegoro, D., Chandran, V., Zhang, J., and Trost, S.G. (2019). Prediction of relative physical activity intensity using multimodal sensing of physiological data. Sensors, 19.","DOI":"10.3390\/s19204509"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Javed, A.R., Sarwar, M.U., Khan, S., Iwendi, C., Mittal, M., and Kumar, N. (2020). Analyzing the effectiveness and contribution of each axis of tri-axial accelerometer sensor for accurate activity recognition. Sensors, 20.","DOI":"10.3390\/s20082216"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Mardini, M.T., Bai, C., Wanigatunga, A.A., Saldana, S., Casanova, R., and Manini, T.M. (2021). Age differences in estimating physical activity by wrist accelerometry using machine learning. Sensors, 21.","DOI":"10.3390\/s21103352"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Capela, N.A., Lemaire, E.D., and Baddour, N. (2015). Feature selection for wearable smartphone-based human activity recognition with able bodied, elderly, and stroke patients. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0124414"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e23681","DOI":"10.2196\/23681","article-title":"The effect of sensor placement and number on physical activity recognition and energy expenditure estimation in older adults: Validation study","volume":"9","author":"Davoudi","year":"2021","journal-title":"JMIR mHealth uHealth"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"954","DOI":"10.1111\/jgs.15331","article-title":"Gait Speed and Mobility Disability: Revisiting Meaningful Levels in Diverse Clinical Populations","volume":"66","author":"Miller","year":"2018","journal-title":"J. Am. Geriatr. Soc."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1419","DOI":"10.1111\/jgs.16524","article-title":"Identification of Sarcopenia Components That Discriminate Slow Walking Speed: A Pooled Data Analysis","volume":"68","author":"Manini","year":"2020","journal-title":"J. Am. Geriatr. Soc."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"572","DOI":"10.1016\/j.gaitpost.2007.07.009","article-title":"Separating the effects of age and walking speed on gait variability","volume":"27","author":"Kang","year":"2008","journal-title":"Gait Posture"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1093\/ageing\/afp250","article-title":"Ageing and gait variability-a population-based study of older people","volume":"39","author":"Callisaya","year":"2010","journal-title":"Age Ageing"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"863","DOI":"10.1177\/0898264312438551","article-title":"Validity and reliability of the short physical performance battery in two diverse older adult populations in Quebec and Brazil","volume":"24","author":"Freire","year":"2012","journal-title":"J. Aging Health"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.conctc.2017.02.003","article-title":"Metabolic costs of daily activity in older adults (Chores XL) study: Design and methods","volume":"6","author":"Corbett","year":"2017","journal-title":"Contemp. Clin. Trials Commun."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"2118","DOI":"10.1111\/j.1532-5415.2011.03655.x","article-title":"Metabolic cost of daily activities and effect of mobility impairment in older adults","volume":"59","author":"Knaggs","year":"2011","journal-title":"J. Am. Geriatr. Soc."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1093\/gerona\/55.4.M221","article-title":"Lower extremity function and subsequent disability: Consistency across studies, predictive models, and value of gait speed alone compared with the short physical performance battery","volume":"55","author":"Guralnik","year":"2000","journal-title":"J. Gerontol. Ser. A Biol. Sci. Med. Sci."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"556","DOI":"10.1056\/NEJM199503023320902","article-title":"Lower-Extremity Function in Persons over the Age of 70 Years as a Predictor of Subsequent Disability","volume":"332","author":"Guralnik","year":"1995","journal-title":"N. Engl. J. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1111\/jgs.15631","article-title":"Effect of Hospitalizations on Physical Activity Patterns in Mobility-Limited Older Adults","volume":"67","author":"Wanigatunga","year":"2019","journal-title":"J. Am. Geriatr. Soc."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Brown, J.D., Lo-Ciganic, W.H., Shao, H., Pahor, M., and Manini, T.M. (2020). Trajectories of short physical performance battery are strongly associated with future major mobility disability: Results from the life study. J. Clin. Med., 9.","DOI":"10.3390\/jcm9082332"},{"key":"ref_26","unstructured":"(2022, April 01). American Time Use Survey Home Page, Available online: https:\/\/www.bls.gov\/tus\/."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"555","DOI":"10.1002\/clc.4960130809","article-title":"Metabolic equivalents (METS) in exercise testing, exercise prescription, and evaluation of functional capacity","volume":"13","author":"Sidney","year":"1990","journal-title":"Clin. Cardiol."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Quiroz, J.C., Banerjee, A., Dascalu, S.M., and Lau, S.L. (2017). Feature Selection for Activity Recognition from Smartphone Accelerometer Data. Intell. Autom. Soft Comput., 1\u20139.","DOI":"10.1080\/10798587.2017.1342400"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1109\/TITB.2005.856864","article-title":"Implementation of a real-time human movement classifier using a triaxial accelerometer for ambulatory monitoring","volume":"10","author":"Karantonis","year":"2006","journal-title":"IEEE Trans. Inf. Technol. Biomed."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1109\/TITB.2007.899496","article-title":"Detection of daily activities and sports with wearable sensors in controlled and uncontrolled conditions","volume":"12","author":"Ermes","year":"2008","journal-title":"IEEE Trans. Inf. Technol. Biomed."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"e11270","DOI":"10.2196\/11270","article-title":"Accuracy of samsung gear s smartwatch for activity recognition: Validation study","volume":"7","author":"Davoudi","year":"2019","journal-title":"JMIR mHealth uHealth"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"396","DOI":"10.1152\/japplphysiol.00026.2015","article-title":"Methods to estimate aspects of physical activity and sedentary behavior from high-frequency wrist accelerometer measurements","volume":"119","author":"Staudenmayer","year":"2015","journal-title":"J. Appl. Physiol."},{"key":"ref_33","first-page":"785","article-title":"XGBoost: A scalable tree boosting system","volume":"Volume 13\u201317","author":"Chen","year":"2016","journal-title":"Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York, NY, USA, 14 August 2016"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v033.i01","article-title":"Regularization paths for generalized linear models via coordinate descent","volume":"33","author":"Friedman","year":"2010","journal-title":"J. Stat. Softw."},{"key":"ref_35","unstructured":"(2022, April 01). Package \u201cglmnet\u201d: Lasso and Elastic-Net Regularized Generalized Linear Models. Available online: https:\/\/cran.r-project.org\/web\/packages\/glmnet\/glmnet.pdf."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Kuhn, M., and Johnson, K. (2013). Applied Predictive Modeling, Springer.","DOI":"10.1007\/978-1-4614-6849-3"},{"key":"ref_37","unstructured":"Kohavi, R. (1995, January 20\u201325). A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the 14th International Joint Conference on Artificial Intelligence, Montreal, QC, Canada."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"3301","DOI":"10.1093\/bioinformatics\/bti499","article-title":"Prediction error estimation: A comparison of resampling methods","volume":"21","author":"Molinaro","year":"2005","journal-title":"Bioinformatics"},{"key":"ref_39","first-page":"2079","article-title":"On over-fitting in model selection and subsequent selection bias in performance evaluation","volume":"11","author":"Cawley","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_40","first-page":"83","article-title":"Statistics The Elements of Statistical Learning","volume":"27","author":"Hastie","year":"2009","journal-title":"Math. Intell."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"841","DOI":"10.2105\/AJPH.82.6.841","article-title":"Functional limitations and disability among elders in the Framingham study","volume":"82","author":"Jette","year":"1992","journal-title":"Am. J. Public Health"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1159\/000089822","article-title":"Comorbidity and physical function: Results from the aging and longevity study in the sirente geographic area (iISIRENTE Study)","volume":"52","author":"Cesari","year":"2006","journal-title":"Gerontology"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1089\/rej.2013.1491","article-title":"Association between short physical performance battery and falls in older people: The progetto veneto anziani study","volume":"17","author":"Veronese","year":"2014","journal-title":"Rejuvenation Res."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1093\/ptj\/pzz154","article-title":"Diagnostic Accuracy of the Short Physical Performance Battery for Detecting Frailty in Older People","volume":"100","author":"Perracini","year":"2020","journal-title":"Phys. Ther."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"1435","DOI":"10.1007\/s40520-018-1082-y","article-title":"Short-Physical Performance Battery (SPPB) score is associated with falls in older outpatients","volume":"31","author":"Lauretani","year":"2019","journal-title":"Aging Clin. Exp. Res."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Lin, B.S., Lee, I.J., Fahn, C.S., Lee, Y.F., Chou, W.J., and Wu, M.L. (2021). Depth-camera based energy expenditure estimation system for physical activity using posture classification algorithm. Sensors, 21.","DOI":"10.3390\/s21124216"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Park, H., Dong, S.Y., Lee, M., and Youn, I. (2017). The role of heart-rate variability parameters in activity recognition and energy-expenditure estimation using wearable sensors. Sensors, 17.","DOI":"10.3390\/s17071698"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1016\/j.medengphy.2019.10.006","article-title":"Prediction of energy expenditure during activities of daily living by a wearable set of inertial sensors","volume":"75","author":"Hedegaard","year":"2020","journal-title":"Med. Eng. Phys."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"2191","DOI":"10.1088\/0967-3334\/35\/11\/2191","article-title":"A random forest classifier for the prediction of energy expenditure and type of physical activity from wrist and hip accelerometers","volume":"35","author":"Ellis","year":"2014","journal-title":"Physiol. Meas."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"206","DOI":"10.1249\/JES.0000000000000206","article-title":"A Framework to evaluate devices that assess physical behavior","volume":"47","author":"Keadle","year":"2019","journal-title":"Exerc. Sport Sci. Rev."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/8\/3061\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T22:55:04Z","timestamp":1760136904000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/8\/3061"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,4,15]]},"references-count":50,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2022,4]]}},"alternative-id":["s22083061"],"URL":"https:\/\/doi.org\/10.3390\/s22083061","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,4,15]]}}}