[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-detail-290493-en":59,"doc-seo-290493-105":82},{"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":5,"data":60},{"doc_id":61,"user_id":62,"nickname":63,"user_avatar":64,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":66,"doc_content":67,"file_id":68,"file_url":69,"file_type":70,"file_size":71,"view_count":72,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":73,"language":74,"language_code":75,"site_id":76,"html_lang":75,"table_of_contents":77,"faqs":78,"seo_title":79,"seo_description":66,"update_tm":80,"read_time":81,"preview_page_count":4},290493,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780","Explainable AI and Machine Learning Towards Human Gait Deterioration Analysis - Abstract","Gait analysis uses non-invasive sensors and machine learning to support healthcare applications, focusing here on detecting cognitive decline in Parkinson's disease under dual-task conditions. Convolutional neural networks combined with explainable machine learning objectively analyze gait data and link model decisions to clinically relevant biomarkers through interpretable mappings to visual human observations and quantitative gait parameters. The method robustly assesses Parkinson’s severity from ground reaction force data, achieving up to 98% F1 on individual datasets and 95.5% on the combined set. It also examines cognitive load in healthy gait for identity verification with 100% F1, and uses Layer-Wise Relevance Propagation to identify weaker predictive features. Results show cognitive deterioration affects balance and foot landing/lifting dynamics across both classification cases.","Explainable AI and Machine Learning Towards Human Gait Deterioration Analysis  \nAbdullah Alharthi, King Abdullah University of Science and Technology (KAUST), Thuwal, [Saudi Arabia. abdullah.alharthi@kaust.edu.sa](Saudi Arabia. abdullah.alharthi@kaust.edu.sa)  \nAbstract  \nGait analysis, an expanding research area, employs non-invasive sensors and machine learning techniques for a range of applications. In this study, we concentrate on gait analysis for detecting cognitive decline in Parkinson's disease (PD) and under dual-task conditions. Using convolutional neural networks (CNNs) and explainable machine learning, we objectively analyze gait data and associate findings with clinically relevant biomarkers. This is accomplished by connecting machine learning outputs to decisions based on human visual observations or derived quantitative gait parameters, which are tested and routinely implemented in current healthcare practice. Our analysis of gait deterioration due to cognitive decline in PD enables robust results using the proposed methods for assessing PD severity from ground reaction force (GRF) data. We achieved classification accuracies of 98% F1 scores for each PhysioNet.org dataset and 95.5% F1 scores for the combined PhysioNet dataset. By linking clinically observable features to the model outputs, we demonstrate the impact of PD severity on gait. Furthermore, we explore the significance of cognitive load in healthy gait analysis, resulting in robust classification accuracies of 100% F1 scores for subject identity verification. We also identify weaker features crucial for model predictions using Layer-Wise Relevance Propagation. A notable finding of this study reveals that cognitive deterioration's effect on gait influences body balance and foot landing/lifting dynamics in both classification cases: cognitive load in healthy gait and cognitive decline in PD gait.  \nKeywords: —Deep convolutional neural networks (DCNN), deep learning, ground reaction forces (GRF), gait, interpretable neural networks, Parkinson’s disease, perturbation.  \n1. Introduction  \nHuman gait is the unique manner in which each individual walks. Gait involves a cyclical sequence of movements of both lower limbs that can be described as a series of transitions between states [1] . There is significant interest in sensing and recognizing human gait for various applications, including healthcare [2], [3], sports [4], [5], biometrics [6],[7],[8], and human-robot interaction [9], [10] . Gait provides  \nimportant information about a person's physical and physiological characteristics, such as weight, gender, health, and age.  \nThe first focus of this work was on gait in Parkinson's disease (PD) patients. PD is a common neurodegenerative disorder caused by loss of neurons in the midbrain [11] . While a definitive PD diagnosis requires pathological examination, limited reliable neuropathological criteria prevent conclusive diagnosis during life. Historically, only 80-90% of clinical PD diagnoses have been confirmed at autopsy [12] . PD is associated with reduced life expectancy, partly due to the 25-40% of patients who develop dementia [11],[12]. Currently, PD diagnosis and severity rating rely primarily on clinical evaluation and subjective surveys using scales like the Unified Parkinson's Disease Rating Scale, Hoehn and Yahr staging, and Schwab and England activities of daily living [13] . Gait deviation is a hallmark of PD, and disease progression increases fall risk [11]. However, in early PD subtle gait changes may lead to inconclusive visual evaluation, partly because slow walking and short strides can also indicate age, mood, or other conditions [12] . Consistent with this, PD manifestsas tremor, rigidity, and bradykinesia. Gait analysis aids PD diagnosis and tracking, though current methods are semi-subjective.  \nThe second focus was on gait under cognitive load (\"dual tasks\"). Gait patterns vary within and between individuals due to factors like dual","cbCaiu8Qh6OBzMVq","https://ap.wps.com/l/cbCaiu8Qh6OBzMVq","pdf",4629759,13,37,"English","en",105,"# Introduction\n## Focus 1: Gait in Parkinson's disease (PD)\n## Focus 2: Gait under cognitive load (dual tasks)\n## Motivation and explainability approach","[{\"question\":\"What is the main goal of this gait analysis study?\",\"answer\":\"To detect cognitive decline in Parkinson's disease under dual-task conditions and use gait under cognitive load for subject identification.\"},{\"question\":\"Which data source and model approach are used?\",\"answer\":\"Ground reaction force (GRF) data are analyzed using convolutional neural networks combined with explainable machine learning methods.\"},{\"question\":\"How does the study interpret the neural network predictions?\",\"answer\":\"Layer-Wise Relevance Propagation (LRP) produces relevance maps that attribute prediction portions to input signals, highlighting important regions and weaker predictive features.\"}]","Explainable AI and Machine Learning Towards Human Gait Deterioration Analysis - Abstract | PDF",1789641832,93,{"code":4,"msg":83,"data":84},"ok",{"site_id":76,"language":75,"slug":85,"title":65,"keywords":86,"description":66,"schema_data":87,"social_meta":141,"head_meta":143,"extra_data":145,"updated_unix":146},"explainable-ai-and-machine-learning-towards-human-gait-deterioration-analysis-abstract","",{"@graph":88,"@context":140},[89,103,123],{"@type":90,"itemListElement":91},"BreadcrumbList",[92,96,98,101],{"item":93,"name":94,"@type":95,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":97,"name":9,"@type":95,"position":14},"https://docshare.wps.com/document/",{"item":99,"name":40,"@type":95,"position":100},"https://docshare.wps.com/document/research-report/",3,{"item":102,"name":65,"@type":95,"position":19},"https://docshare.wps.com/document/explainable-ai-and-machine-learning-towards-human-gait-deterioration-analysis-abstract/290493/",{"url":102,"name":65,"@type":104,"image":105,"author":110,"headline":65,"publisher":112,"fileFormat":115,"inLanguage":75,"description":66,"dateModified":116,"datePublished":117,"encodingFormat":115,"isAccessibleForFree":118,"interactionStatistic":119},"DigitalDocument",{"url":106,"@type":107,"width":108,"height":109},"https://docshare.wps.com/thumbnails/explainable-ai-and-machine-learning-towards-human-gait-deterioration-analysis-abstract/290493.png","ImageObject",300,407,{"name":63,"@type":111},"Person",{"url":93,"name":113,"@type":114},"DocShare","Organization","application/pdf","2026-10-21","2026-09-17",true,{"@type":120,"interactionType":121,"userInteractionCount":72},"InteractionCounter",{"@type":122},"ViewAction",{"@type":124,"mainEntity":125},"FAQPage",[126,132,136],{"name":127,"@type":128,"acceptedAnswer":129},"What is the main goal of this gait analysis study?","Question",{"text":130,"@type":131},"To detect cognitive decline in Parkinson's disease under dual-task conditions and use gait under cognitive load for subject identification.","Answer",{"name":133,"@type":128,"acceptedAnswer":134},"Which data source and model approach are used?",{"text":135,"@type":131},"Ground reaction force (GRF) data are analyzed using convolutional neural networks combined with explainable machine learning methods.",{"name":137,"@type":128,"acceptedAnswer":138},"How does the study interpret the neural network predictions?",{"text":139,"@type":131},"Layer-Wise Relevance Propagation (LRP) produces relevance maps that attribute prediction portions to input signals, highlighting important regions and weaker predictive features.","https://schema.org",{"og:url":102,"og:type":142,"og:title":65,"og:site_name":113,"og:description":66},"article",{"robots":144,"canonical":102},"index,follow",{"doc_id":61,"site_id":76},1790463795]