[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-127097-105":59,"doc-detail-127097-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","machine-learning-approaches-for-identification-of-parkinsons-disease-severity-using-multimodal-features","MACHINE LEARNING APPROACHES FOR IDENTIFICATION OF PARKINSON’S DISEASE SEVERITY USING MULTIMODAL FEATURES","","Machine learning models were developed to create a digital biomarker for assessing Parkinson’s disease (PD) severity. The study included 50 PD patients and 50 healthy controls who completed clinical tests, gait analysis via a motion capture system, postural and functional evaluations, and lifestyle questionnaires. Multimodal features underwent dimensionality reduction using logistic regression and principal component analysis, then severity was identified using the MDS-UPDRS total score. Six models were trained with linear regression and random forest, with Model 1 performing best; gait spatiotemporal variables were most informative.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/machine-learning-approaches-for-identification-of-parkinsons-disease-severity-using-multimodal-features/127097/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/machine-learning-approaches-for-identification-of-parkinsons-disease-severity-using-multimodal-features/127097.png","ImageObject",300,407,{"name":92,"@type":93},"Levi","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-08-05",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is the study’s main goal for Parkinson’s disease patients?","Question",{"text":113,"@type":114},"To build a digital biomarker that estimates PD severity using multimodal features and machine learning based on the MDS-UPDRS total score.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"Which data types were used as multimodal inputs?",{"text":118,"@type":114},"Clinical tests, motion-capture gait analysis, postural and functional assessments, and lifestyle questionnaires were collected from both PD patients and healthy controls.",{"name":120,"@type":111,"acceptedAnswer":121},"How was PD severity prediction modeled and evaluated?",{"text":122,"@type":114},"Multimodal features were reduced in dimensionality using logistic regression and principal component analysis, then six machine-learning models (linear regression and random forest) were trained, with spatiotemporal gait variables proving most important.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},127097,1785936830,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":19,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":130,"read_time":52},5909887256941,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","MACHINE LEARNING APPROACHES FOR IDENTIFICATION OF PARKINSON’S DISEASE SEVERITY USING MULTIMODAL FEATURES  \nHwayoung Park1, Changhong Youm1,2*, Bohyun Kim1,2, Hyejin Choi1,2, Juseon  \nHwang1,2, and Minsoo Kim1,2  \n1 Biomechanics laboratory, Dong-A University, Busan, Republic of Korea  \n2 Department of Health Sciences, The Graduate School of Dong-A University,  \nBusan, Republic of Korea  \nThis study aimed to create a digital biomarker for assessing the severity of Parkinson's  \ndisease (PD) using multimodal features from 50 PD patients and 50 healthy controls. They  \nunderwent clinical tests, gait analysis using a motion capture system, postural and  \nfunctional evaluations, and lifestyle questionnaires. These multimodal features underwent  \ndimensionality reduction techniques such as logistic regression and principal component  \nanalysis to identify PD severity using the MDS-UPDRS total score. The results developed  \nsix models using machine learning algorithms (Linear Regression and Random Forest),  \nwith Model 1 performing the best; spatiotemporal variables from gait analysis were crucial  \nin identifying PD severity. We aim to identify important features correlated with MDS  \nUPDRS and expect to be applied in clinical settings to monitor the severity of PD.  \nKEYWORDS: Parkinson’s disease, symptom severity , kinematics, multimodal, machine  \nlearning  \nINTRODUCTION: Evaluating the severity of Parkinson's disease (PD) is crucial but challenging due to the subjective nature of current methods reliant on clinical scales such asthe Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS)(Sotirakis et al. , 2023) . Limitations include interrater discrepancies hindering disease progression detection and restrictive statistical analyses in clinical research (Almgren et al. , 2023) . Treatment methods predominantly aim at symptom relief with medications such as levodopa (Albrecht et al. , 2022) . Leveraging digital biomarkers addresses these limitations by providing personalized therapy and novel intervention targets (Birkenbihl et al. , 2023; Deng et al. , 2022) . The key PD features, such as tremors and gait impairments, can be converted into quantitative surrogates aiding symptom evaluation remotely, minimizing disruptions in daily life  \n(Deng et al. , 2022; Sotirakis et al. , 2023) .  \nRecent studies explored multimodal datasets encompassing clinical and movement data, voice recordings, magnetic resonance imaging (MRI), and handwriting patterns to identify PD symptom severity across different neurological states (Sotirakis et al. , 2023) . Among these multimodal features, physical, functional, and kinematic data could be used to identify symptom severity in the early and advanced stages of PD (Deng et al. , 2022) . Several studies have developed PD severity identification and diagnostic models using machine learning (ML) and deep learning techniques with these digital biomarkers (Albrecht et al. , 2022; Almgren et al. , 2023; Birkenbihl et al. , 2023; Deng et al. , 2022; Sotirakis et al. , 2023) . However, analyzing extensive datasets generated by digital devices necessitates feature-reduction steps to extract  \nclinically meaningful information while avoiding errors (Sotirakis et al. , 2023) .  \nIn previous research, ML algorithms applied to data from PD clinical questionnaires and wearable devices effectively distinguish between healthy individuals and those with PD or similar disorders. (Almgren et al. , 2023; Birkenbihl et al. , 2023; Deng et al. , 2022; Sotirakis et al. , 2023) . These algorithms could identify individuals with freezing episodes and falls, as well as detect various symptom signs in individuals with PD. These studies have shown that combining multimodal data collected from individuals with PD and ML algorithms could more effectively track and predict symptom progression than traditional clinical assessment scales (Albrecht et al. , 2022; Almgren et al. , 2023) . However, despite many s","cbCaieNV4tLddS6P","https://ap.wps.com/l/cbCaieNV4tLddS6P","pdf",211073,"English","# Introduction\n# Methods\n## Participants and study design\n## Clinical and multimodal assessments\n## Feature processing and model training","[{\"question\":\"What is the study’s main goal for Parkinson’s disease patients?\",\"answer\":\"To build a digital biomarker that estimates PD severity using multimodal features and machine learning based on the MDS-UPDRS total score.\"},{\"question\":\"Which data types were used as multimodal inputs?\",\"answer\":\"Clinical tests, motion-capture gait analysis, postural and functional assessments, and lifestyle questionnaires were collected from both PD patients and healthy controls.\"},{\"question\":\"How was PD severity prediction modeled and evaluated?\",\"answer\":\"Multimodal features were reduced in dimensionality using logistic regression and principal component analysis, then six machine-learning models (linear regression and random forest) were trained, with spatiotemporal gait variables proving most important.\"}]","MACHINE LEARNING APPROACHES FOR IDENTIFICATION OF PARKINSON’S DISEASE SEVERITY USING MULTIMODAL FEATURES | PDF"]