[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119856-en":3,"doc-seo-119856-105":30,"detail-sidebar-cat-0-en-105":94},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119856,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Estimating Lower Limb Joint Moments in Gait Using Common Machine Learning Approaches","Aim of the study was to evaluate whether common machine learning algorithms can estimate lower-limb joint moments during fast walking gait. Kinematic and ground reaction force data from 19 participants were collected using a motion capture system and force plates. Inverse dynamics generated reference hip, knee, and ankle moments, while models including Random Forest, Linear Regression, Neural Network, AdaBoost, and Gradient Boosting predicted moments from kinematic data only. High R² values (>0.9) were reported for ankle, knee, and hip across anatomical planes for several models, supporting clinical feasibility for injury prevention monitoring.","ESTIMATING LOWER LIMB JOINT MOMENTS IN GAIT USING COMMON MACHINE  \nLEARNING APPROACHES  \nAlex Ong1 and Joseph Hamill2  \nSchool of Sports, Health and Leisure, Republic Polytechnic, Singapore 1 School of Public Health & Health Sciences, University of Massachusetts  \nAmherst, USA2  \nThe aim of this study was to investigate the efficacy of common machine learning algorithmic approaches to estimate lower limb joint moments during fast walking gait. Kinematic and ground reaction force data on 19 participants were captured with a force-plate and motion caption capture system. Inverse dynamics was used to calculate the right lower limb joint moments and common machine learning algorithmic approaches , such as Random Forest (RF), Linear Regression (LR), Neural Network (NN), AdaBoost (AB) and Gradient Boosting, were used to predict the corresponding joint moments using only the kinematic data. High coefficient of determination values (R2>0 .9) for predicting moments using random forest , neural network and AdaBoost are observed in for the ankle, knee and hip joints in frontal, sagittal and transverse planes. The other approaches had R2 values between ranged 0.71 and 0.97. This suggests that common machine learning algorithms may be a feasible approach to estimate joint moments during fast walking in a clinical setting for  \nmonitoring sport injury prevention and management.  \nKEYWORDS: data science, machine learning, joint moments, fast walking, gait  \nINTRODUCTION: The measurement of joint moments during gait traditionally using forcetransducers instrumentation is costly and potentially limits its adoption (Hong et al. , 2017) . There are studies that indirectly estimate joint moments during gait without the use of force transducers (Rokhmanova et al. , 2022; Johnson et al. , 2018; Onal et al. , 2019) . The use of machine learning in data science for such estimations is an emerging area of investigation (Burdack et al. , 2019) .  \nHowever, this type of investigation typically requires ‘large data’ to optimize ecological validity (Ferber et al. , 2016) . Recently, machine learning approaches have been applied to small or minimal data set that require models with low complexity to avoid overfitting the model to the data (Rokhmanova et al. , 2022; Lim et al. , 2020) . Commonly used machine learning (ML) approaches are Random Forest (RF), Linear Regression (LR), Neural Network (NN), AdaBoost (AB) and Gradient Boosting (GB) (Donisi et al. , 2021) .  \nIn laboratory-based gait measurements of the lower limb , the motion is constrained within a small spatial volume and the activities are repetitive (Ong et al. , 2017) . This is potentially advantageous in a simple model, where faster computation time with fewer relevant features/predictors are required during the modelling process, as compared to a complex movement model (e.g. , dancing and cutting maneuvers) . However, there is a dearth of knowledge on investigating the efficacy of estimating or predicting joint moments using commonly used machine learning algorithmic approaches.  \nTherefore, the aim of this pilot study was to investigate the efficacy of common machine learning algorithmic approaches to estimate joint moments in overground fast walking. It was hypothesized that the predicted lower limb joint moments would have good correspondence with the measured for joint moment in the three orthogonal anatomical planes using common machine learning algorithmic approaches.  \nMETHODS: Nineteen healthy participants (age = 33.78 + 6.20 yr, height = 1.60 + 0.06 m, mass = 55.56 + 7.56 kg) were included in the study. The study was approved by the institutional review board and written consent was obtained from the participants prior to the study.  \nKinematic variables during walking were collected using a six-camera, motion capture system  \nPublished by NMU Commons, 2023 1  \n(Motion Analysis Corporation, California) sampling at 100 Hz. The ground reaction force data were collected from two floor-","cbCaie7wz3ASmpGb","https://ap.wps.com/l/cbCaie7wz3ASmpGb","pdf",468301,1,4,"English","en",105,"# Introduction\n## Background and motivation\n## Study aim and hypothesis\n# Methods\n## Participants and ethics\n## Data collection and preprocessing\n## Model training and evaluation\n# Results\n## Walking speed and predictive performance","[{\"question\":\"What data sources and methods were used to estimate the reference joint moments?\",\"answer\":\"Kinematic data were captured with a motion capture system and ground reaction forces were collected using force plates. Inverse dynamics calculated hip, knee, and ankle joint moments for the right lower limb as reference values.\"},{\"question\":\"Which machine learning algorithms were compared for predicting joint moments from kinematics only?\",\"answer\":\"Random Forest, Linear Regression, Neural Network, AdaBoost, and Gradient Boosting were trained to predict corresponding joint moments using only kinematic data.\"},{\"question\":\"How effective were the machine learning models for fast walking gait?\",\"answer\":\"Several models achieved high coefficient of determination values (R² \\u003e 0.9) for ankle, knee, and hip joint moments across frontal, sagittal, and transverse planes, while other approaches showed R² values in a lower but still substantial range (about 0.71 to 0.97).\"},{\"question\":\"Why is this approach relevant for clinical use and injury prevention?\",\"answer\":\"The study suggests common machine learning algorithms can feasibly estimate joint moments during fast walking in a clinical setting, enabling monitoring for sport injury prevention and management without relying on costly force-transducer instrumentation.\"}]","Estimating Lower Limb Joint Moments in Gait Using Common Machine Learning Approaches | PDF",1785726676,10,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":89,"head_meta":91,"extra_data":93,"updated_unix":28},"estimating-lower-limb-joint-moments-in-gait-using-common-machine-learning-approaches","",{"@graph":36,"@context":88},[37,53,67],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":21},"https://docshare.wps.com/document/estimating-lower-limb-joint-moments-in-gait-using-common-machine-learning-approaches/119856/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80,84],{"name":71,"@type":72,"acceptedAnswer":73},"What data sources and methods were used to estimate the reference joint moments?","Question",{"text":74,"@type":75},"Kinematic data were captured with a motion capture system and ground reaction forces were collected using force plates. Inverse dynamics calculated hip, knee, and ankle joint moments for the right lower limb as reference values.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"Which machine learning algorithms were compared for predicting joint moments from kinematics only?",{"text":79,"@type":75},"Random Forest, Linear Regression, Neural Network, AdaBoost, and Gradient Boosting were trained to predict corresponding joint moments using only kinematic data.",{"name":81,"@type":72,"acceptedAnswer":82},"How effective were the machine learning models for fast walking gait?",{"text":83,"@type":75},"Several models achieved high coefficient of determination values (R² > 0.9) for ankle, knee, and hip joint moments across frontal, sagittal, and transverse planes, while other approaches showed R² values in a lower but still substantial range (about 0.71 to 0.97).",{"name":85,"@type":72,"acceptedAnswer":86},"Why is this approach relevant for clinical use and injury prevention?",{"text":87,"@type":75},"The study suggests common machine learning algorithms can feasibly estimate joint moments during fast walking in a clinical setting, enabling monitoring for sport injury prevention and management without relying on costly force-transducer instrumentation.","https://schema.org",{"og:url":52,"og:type":90,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":92,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":95},[96,100,104,108,113,118,123,126,131,134,137],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":97,"show_sort_weight":98,"slug":99},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":101,"show_sort_weight":102,"slug":103},"Literature",80,"literature",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":105,"show_sort_weight":106,"slug":107},"Exam",70,"exam",{"id":109,"doc_module":4,"doc_module_name":46,"category_name":110,"show_sort_weight":111,"slug":112},5,"Comic",60,"comic",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},6,"Technology",50,"technology",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":124,"slug":125},30,"research-report",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":129,"slug":130},9,"Religion & Spirituality",20,"religion-spirituality",{"id":129,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":129,"slug":133},"World Cup","world-cup",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":29,"slug":136},"Lifestyle","lifestyle",{"id":138,"doc_module":4,"doc_module_name":46,"category_name":139,"show_sort_weight":109,"slug":140},19,"General","general"]