{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T18:31:41Z","timestamp":1788546701228,"version":"build-2803163510"},"reference-count":28,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Mhealth Uhealth"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Single-leg squat (SLS) performance is widely used to screen functional movement quality, but practical assessment often relies on expert visual grading or laboratory-based motion capture. In addition, conventional SLS criteria mainly focus on isolated joint deviations and may overlook coordination-related, multisegment movement patterns that characterize impaired performance.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to examine the feasibility of an interpretable machine learning framework for classifying SLS performance into 3 levels (good, moderate, and poor) from single-smartphone, frontal-view videos based on trunk, pelvic, and knee kinematics, and to evaluate coordination-informed features and model explainability using Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME).<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n A dataset of frontal-view SLS videos was labeled by physiotherapists into 3 functional categories (good, moderate, and poor). Videos were processed using 2D pose estimation, and models were trained on 17 engineered kinematic features derived from trunk, pelvic, and knee angles. Following the feature selection, 7 classifiers were trained and evaluated using the 8 selected features with stratified 5-fold cross-validation and a held-out test set. SHAP and LIME were applied for global and local interpretability.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n On the held-out test set, adaptive boosting classified SLS performance with an accuracy of 0.84, an\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 0.85, and an area under curve of 0.92. SHAP indicated that the summated angle (trunk + pelvis + knee), coordination-related features (knee \u00d7 trunk interaction and knee-to-trunk ratio), and knee angle were key contributors to model predictions. LIME provided instance-level explanations that helped interpret individual classifications and decision boundaries.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n This study presents an interpretable machine learning framework for classifying SLS performance into 3 levels using frontal-view videos acquired with a single smartphone. By leveraging coordination-informed engineered features and explainable artificial intelligence, the framework enables transparent interpretation of movement performance beyond isolated joint deviations. The proposed workflow uses smartphones for standardized video acquisition, while performance screening is achieved through machine learning. Given its lightweight feature design, this framework has potential for future implementation on modern smartphones and may support rehabilitation planning and injury-prevention strategies in sports and clinical settings.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/85126","type":"journal-article","created":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T16:35:06Z","timestamp":1770309306000},"page":"e85126-e85126","source":"Crossref","is-referenced-by-count":1,"title":["Smartphone-Based Interpretable Machine Learning for Classifying Single-Leg Squat Performance Using Trunk, Pelvic, and Knee Kinematics: Cross-Sectional Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1870-2767","authenticated-orcid":false,"given":"Sihyun","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3521-3121","authenticated-orcid":false,"given":"Kyuenam","family":"Park","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,3,12]]},"reference":[{"issue":"16","key":"R1","doi-asserted-by":"publisher","first-page":"1568","DOI":"10.1001\/jama.2023.19675","article-title":"Evaluation and treatment of knee pain: a review","volume":"330","author":"Duong","journal-title":"JAMA"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","DOI":"10.3390\/muscles4010003","article-title":"The influence of knee proprioception and strength on lower-limb functional symmetry in healthy adults","volume":"4","author":"Drigny","journal-title":"Muscles"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","DOI":"10.1177\/23259671211068438","article-title":"Lower extremity injury after return to sports from concussion: a systematic review","volume":"10","author":"Jildeh","journal-title":"Orthop J Sports Med"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.20218","article-title":"Comparing the reliability of the single leg squat test using two, three, and four category ordinal rating scales","volume":"13","author":"Zhang","journal-title":"PeerJ"},{"issue":"1","key":"R5","doi-asserted-by":"publisher","DOI":"10.1136\/bmjsem-2019-000541","article-title":"Visual assessment of movement quality in the single leg squat test: a review and meta-analysis of inter-rater and intrarater reliability","volume":"5","author":"Ressman","journal-title":"BMJ Open Sport Exerc Med"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","first-page":"33","DOI":"10.46497\/ArchRheumatol.2024.9814","article-title":"Three-dimensional kinematics of the trunk, pelvis, hip, and knee during the single-leg squat and hip torque in subjects with isolated patellofemoral osteoarthritis compared to individually matched controls: preliminary results","volume":"39","author":"Carvalho","journal-title":"Arch Rheumatol"},{"issue":"1","key":"R7","doi-asserted-by":"publisher","first-page":"17","DOI":"10.2519\/josptmethods.2024.0086","article-title":"Evaluation of the ability of machine learning-models to assess postural orientation errors during a single-leg squat","volume":"1","author":"Nae","journal-title":"JOSPT Methods"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3316009","article-title":"AI trainer: autoencoder based approach for squat analysis and correction","volume":"11","author":"Chariar","journal-title":"IEEE Access"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.1109\/JTEHM.2017.2736559","article-title":"Automated assessment of dynamic knee valgus and risk of knee injury during the single leg squat","volume":"5","author":"Kianifar","journal-title":"IEEE J Transl Eng Health Med"},{"issue":"2","key":"R10","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1123\/jsr.2024-0182","article-title":"Markerless kinematic data in the frontal plane contributions to movement quality in the single-leg squat test: a comparison and decision tree approach","volume":"34","author":"Park","journal-title":"J Sport Rehabil"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2020.104021","article-title":"Automated classification of movement quality using the Microsoft Kinect V2 sensor","volume":"125","author":"Dajime","journal-title":"Comput Biol Med"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.109338","article-title":"Explaining deep learning models for age-related gait classification based on acceleration time series","volume":"184","author":"Zheng","journal-title":"Comput Biol Med"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.105329","article-title":"Association between central sensitization and gait in chronic low back pain: insights from a machine learning approach","volume":"144","author":"Zheng","journal-title":"Comput Biol Med"},{"issue":"16","key":"R14","doi-asserted-by":"publisher","DOI":"10.3390\/s20164539","article-title":"Validity of a new 3-D motion analysis tool for the assessment of knee, hip and spine joint angles during the single leg squat","volume":"20","author":"Tak","journal-title":"Sensors (Basel)"},{"issue":"6","key":"R15","doi-asserted-by":"publisher","first-page":"100857","DOI":"10.1016\/j.landig.2025.01.013","article-title":"Importance of sample size on the quality and utility of AI-based prediction models for healthcare","volume":"7","author":"Riley","journal-title":"Lancet Digit Health"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2025.112738","article-title":"Artificial intelligence-enhanced 3D gait analysis with a single consumer-grade camera","volume":"187","author":"Guo","journal-title":"J Biomech"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.gaitpost.2022.05.020","article-title":"The influence of task type and movement speed on lower limb kinematics during single-leg tasks","volume":"96","author":"Rabello","journal-title":"Gait Posture"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.gaitpost.2018.02.016","article-title":"Single leg squat ratings by clinicians are reliable and predict excessive hip internal rotation moment","volume":"61","author":"Barker-Davies","journal-title":"Gait Posture"},{"issue":"1","key":"R19","doi-asserted-by":"publisher","DOI":"10.1186\/s13102-021-00289-x","article-title":"Visual assessment of movement quality: a study on intra- and interrater reliability of a multi-segmental single leg squat test","volume":"13","author":"Ressman","journal-title":"BMC Sports Sci Med Rehabil"},{"issue":"7","key":"R20","first-page":"862","volume":"9","author":"McKeown","journal-title":"Int J Sports Phys Ther"},{"issue":"6","key":"R21","doi-asserted-by":"publisher","first-page":"478","DOI":"10.1097\/JSM.0000000000000288","article-title":"Single-leg squat as a tool to evaluate young athletes' frontal plane knee control","volume":"26","author":"R\u00e4is\u00e4nen","journal-title":"Clin J Sport Med"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","first-page":"2450026","DOI":"10.1142\/S021951942450026X","article-title":"Single-leg squat postural sway reliability: how many trials to analyze for chronic low back pain?","volume":"25","author":"Kwok","journal-title":"J Mech Med Biol"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.3390\/s23010003","article-title":"Validation of angle estimation based on body tracking data from RGB-D and RGB cameras for biomechanical assessment","volume":"23","author":"Lafayette","journal-title":"Sensors (Basel)"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2025.3530840","article-title":"XAI-enhanced machine learning for obesity risk classification: a stacking approach with LIME explanations","volume":"13","author":"Azad","journal-title":"IEEE Access"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.3389\/fnagi.2023.1267020","article-title":"An explainable machine learning based prediction model for Alzheimer\u2019s disease in China longitudinal aging study","volume":"15","author":"Yue","journal-title":"Front Aging Neurosci"},{"issue":"Pt A","key":"R26","doi-asserted-by":"publisher","first-page":"110962","DOI":"10.1016\/j.compbiomed.2025.110962","article-title":"Squat errors classification based on National Academy of Sports Medicine guidelines using IMU and deep learning algorithms","volume":"197","author":"Khalilipour","journal-title":"Comput Biol Med"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","DOI":"10.1038\/s41698-024-00553-6","article-title":"All models are wrong and yours are useless: making clinical prediction models impactful for patients","volume":"8","author":"Markowetz","journal-title":"NPJ Precis Oncol"},{"issue":"1","key":"R28","doi-asserted-by":"publisher","first-page":"58","DOI":"10.26603\/ijspt20180058","article-title":"Trunk lean during a single-leg squat is associated with trunk lean during pitching","volume":"13","author":"Plummer","journal-title":"Int J Sports Phys Ther"}],"container-title":["JMIR mHealth and uHealth"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/mhealth.jmir.org\/2026\/1\/e85126","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:19:12Z","timestamp":1782843552000},"score":1,"resource":{"primary":{"URL":"https:\/\/mhealth.jmir.org\/2026\/1\/e85126"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,12]]},"references-count":28,"URL":"https:\/\/doi.org\/10.2196\/85126","relation":{},"ISSN":["2291-5222"],"issn-type":[{"value":"2291-5222","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,12]]},"article-number":"v14i16e85126"}}