{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T13:23:17Z","timestamp":1787318597785,"version":"build-2736575974"},"reference-count":24,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"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. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Bradykinesia, a primary symptom of Parkinson\u2019s disease, significantly impacts patients\u2019 quality of life. Traditional assessments like the MDS-UPDRS require in-person evaluation and can be subjective, resulting in patient burden and variability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study investigates digital health technologies and machine learning (ML) models, specifically XGBoost and Convolutional Neural Networks (CNNs), to improve objectivity and precision in bradykinesia quantification. Using data from the 12-month WATCH-PD study in early, untreated Parkinson\u2019s disease, models were evaluated against MDS-UPDRS items (3.6 pronation-supination and 3.7 toe tapping) for cross-sectional accuracy and longitudinal sensitivity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Results show that the class-balanced XGBoost was the best-performing model on both tasks, outperforming a stratified-random baseline and all CNN variants (cross-entropy, focal loss, and self-attention) across every imbalance-aware metric: macro-F1 0.47 and 0.40, balanced accuracy 0.47 and 0.42, quadratic weighted kappa 0.58 and 0.47, and accuracy 0.60 and 0.53 for pronation-supination and toe tapping, respectively, with the clearest gains on the under-represented higher-severity classes. A secondary longitudinal analysis found limited sensitivity to 12-month change across the ML models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Overall, machine-learning models, led by the class-balanced XGBoost, show promise in accurately predicting item-level MDS-UPDRS bradykinesia scores from wearable sensor data. Our results also highlight the importance of aligning feature selection and model architecture to intended clinical endpoints.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1854818","type":"journal-article","created":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T12:24:31Z","timestamp":1787315071000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparison of machine learning models for bradykinesia assessment using digital health technology"],"prefix":"10.3389","volume":"8","author":[{"given":"Matthew D.","family":"Czech","sequence":"first","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Deng","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Josh","family":"Cosman","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michelle","family":"Crouthamel","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sheng","family":"Zhong","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li","family":"Wang","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"E. Ray","family":"Dorsey","sequence":"additional","affiliation":[{"name":"Department of Neurology, University of Rochester Medical Center","place":["Rochester, NY, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jamie L.","family":"Adams","sequence":"additional","affiliation":[{"name":"Department of Neurology, University of Rochester Medical Center","place":["Rochester, NY, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Shen","sequence":"additional","affiliation":[{"name":"Data and Statistical Sciences, AbbVie","place":["North Chicago, IL, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"368","DOI":"10.1136\/jnnp.2007.131045","article-title":"Parkinson\u2019s disease: clinical features and diagnosis","volume":"79","author":"Jankovic","year":"2008","journal-title":"J Neurol Neurosurg Psychiatry"},{"key":"B2","doi-asserted-by":"publisher","first-page":"2129","DOI":"10.1002\/mds.22340","article-title":"Movement disorder society-sponsored revision of the unified Parkinson\u2019s disease rating scale (MDS-UPDRS): scale presentation and clinimetric testing results","volume":"23","author":"Goetz","year":"2008","journal-title":"Mov Disord"},{"key":"B3","doi-asserted-by":"publisher","first-page":"10189","DOI":"10.3390\/app142210189","article-title":"Evaluating motor symptoms in parkinson\u2019s disease through wearable sensors: a systematic review of digital biomarkers","volume":"14","author":"Polvorinos-Fern\u00e1ndez","year":"2024","journal-title":"Appl Sci"},{"key":"B4","doi-asserted-by":"publisher","first-page":"783","DOI":"10.1007\/s00702-022-02575-5","article-title":"The use of wearables for the diagnosis and treatment of Parkinson\u2019s disease","volume":"130","author":"Reichmann","year":"2023","journal-title":"J Neural Transm"},{"key":"B5","doi-asserted-by":"publisher","first-page":"3879","DOI":"10.3390\/electronics11233879","article-title":"Bradykinesia detection in Parkinson\u2019s disease using smartwatches\u2019 inertial sensors and deep learning methods","volume":"11","author":"Sigcha","year":"2022","journal-title":"Electronics"},{"key":"B6","doi-asserted-by":"publisher","first-page":"5589","DOI":"10.3390\/s21165589","article-title":"Review of wearable devices and data collection considerations for connected health","volume":"21","author":"Vijayan","year":"2021","journal-title":"Sensors"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1016","DOI":"10.1109\/TBME.2015.2480242","article-title":"A method for automatic and objective scoring of bradykinesia using orientation sensors and classification algorithms","volume":"63","author":"Martinez-Manzanera","year":"2016","journal-title":"IEEE Trans Biomed Eng"},{"key":"B8","doi-asserted-by":"publisher","first-page":"2614","DOI":"10.1007\/s10439-017-1908-3","article-title":"Objective assessment of bradykinesia estimated from the wrist extension in older adults and patients with Parkinson\u2019s disease","volume":"45","author":"Rabelo","year":"2017","journal-title":"Ann Biomed Eng"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1476","DOI":"10.3390\/s22041476","article-title":"Deep learning in human activity recognition with wearable sensors: a review on advances","volume":"22","author":"Zhang","year":"2022","journal-title":"Sensors"},{"key":"B10","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1038\/s41746-024-01136-2","article-title":"A wearable sensor and machine learning estimate step length in older adults and patients with neurological disorders","volume":"7","author":"Zadka","year":"2024","journal-title":"npj Digit Med"},{"key":"B11","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1038\/s41531-024-00721-2","article-title":"Using a smartwatch and smartphone to assess early Parkinson\u2019s disease in the watch-pd study over 12 months","volume":"10","author":"Adams","year":"2024","journal-title":"npj Park Dis"},{"key":"B12","doi-asserted-by":"publisher","first-page":"64","DOI":"10.1038\/s41531-023-00497-x","article-title":"Using a smartwatch and smartphone to assess early Parkinson\u2019s disease in the watch-pd study","volume":"9","author":"Adams","year":"2023","journal-title":"npj Park Dis"},{"key":"B13","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1038\/s43856-024-00481-3","article-title":"Improved measurement of disease progression in people living with early Parkinson\u2019s disease using digital health technologies","volume":"4","author":"Czech","year":"2024","journal-title":"Commun Med"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1364","DOI":"10.2522\/ptj.20150580","article-title":"Balance and gait represent independent domains of mobility in Parkinson disease","volume":"96","author":"Horak","year":"2016","journal-title":"Phys Ther"},{"key":"B15","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"B16","doi-asserted-by":"publisher","first-page":"189","DOI":"10.4258\/hir.2021.27.3.189","article-title":"Impact of the choice of cross-validation techniques on the results of machine learning-based diagnostic applications","volume":"27","author":"Tougui","year":"2021","journal-title":"Healthc Inform Res"},{"key":"B17","doi-asserted-by":"publisher","first-page":"235","DOI":"10.1016\/j.eswa.2016.04.032","article-title":"Human activity recognition with smartphone sensors using deep learning neural networks","volume":"59","author":"Ronao","year":"2016","journal-title":"Expert Syst Appl"},{"key":"B18","author":"Yang","year":""},{"key":"B19","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"B20","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1109\/JBHI.2016.2636456","article-title":"Sensor-based gait parameter extraction with deep convolutional neural networks","volume":"21","author":"Hannink","year":"2017","journal-title":"IEEE J Biomed Health Inform"},{"key":"B21","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1038\/s41591-018-0268-3","article-title":"Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network","volume":"25","author":"Hannun","year":"2019","journal-title":"Nat Med"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.neucom.2018.03.032","article-title":"A hybrid spatio-temporal model for detection and severity rating of Parkinson\u2019s disease from gait data","volume":"315","author":"Zhao","year":"2018","journal-title":"Neurocomputing"},{"key":"B23","doi-asserted-by":"publisher","first-page":"105","DOI":"10.1056\/NEJMp1611252","article-title":"Patient-reported outcomes-harnessing patients\u2019 voices to improve clinical care","volume":"376","author":"Basch","year":"2017","journal-title":"N Engl J Med"},{"key":"B24","year":""}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1854818\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T12:24:31Z","timestamp":1787315071000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1854818\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":24,"alternative-id":["10.3389\/fdgth.2026.1854818"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1854818","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]},"article-number":"1854818"}}