A Hybrid Unsupervised Learning Framework for Safety Analysis of Conditionally Automated Vehicle Use in Older Adults with Cognitive Challenges

A Hybrid Unsupervised Learning Framework for Safety Analysis of Conditionally Automated Vehicle Use in Older Adults with Cognitive Challenges | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A Hybrid Unsupervised Learning Framework for Safety Analysis of Conditionally Automated Vehicle Use in Older Adults with Cognitive Challenges Gelareh Hajian, Bing Ye, Shabnam Haghzare, Mark Rapoport, Gary Naglie, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7358601/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 4 You are reading this latest preprint version Abstract Driving is a critical aspect of independence and quality of life, particularly for older adults. However, cognitive challenges such as Mild Cognitive Impairment (MCI), and dementia often lead to driving cessation, resulting in negative social, emotional, and health outcomes. Conditionally Automated vehicles (CAVs) may offer opportunities as an assistive technology to extend driving years for older adults. Despite advancements in CAV technologies, significant gaps remain in ensuring safety during takeover requests (TORs) for individuals with cognitive challenges, as no studies have evaluated their performance using CAVs. Moreover, existing studies analyzing driving performance in CAVs largely focus on comparing isolated metrics such as reaction time (RT) or takeover quality (TOQ) metrics, lacking an integrated framework to assess overall safety, representing another critical research gap. In this study, we address these gaps by developing a novel hybrid unsupervised learning framework to comprehensively analyze driving performance during TORs in older adults with cognitive challenges. Using data from 37 cognitively healthy older adults, we first apply Kmeans clustering combined with domain knowledge to cluster cognitively healthy participants into “safe” and “less safe” driving profiles, establishing a standardized safety baseline. This framework is then extended to evaluate CAV safety among older adults with cognitive challenges (Subjective Cognitive Decline (SCD), MCI, and dementia). To validate the robustness of our approach, we compare the hybrid framework against traditional clustering methods (Kmeans, hierarchical, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN)). Results demonstrate that the proposed hybrid framework outperforms traditional methods, providing a more comprehensive and individualized evaluation of CAV safety during TORs. This study highlights the importance of integrating domain knowledge with data-driven methods for developing a safety framework, paving the way for CAVs safety analysis to support older adults’ independence and mobility. Automated vehicles Older adults with cognitive impairment Driving performance Safety analysis Machine Learning Unsupervised learning Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 29 Aug, 2025 Editor assigned by journal 27 Aug, 2025 Submission checks completed at journal 27 Aug, 2025 First submitted to journal 12 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7358601","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":499564774,"identity":"b5666148-8279-4f0b-8aae-5ffd5eb825fc","order_by":0,"name":"Gelareh Hajian","email":"data:image/png;base64,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","orcid":"","institution":"University of Toronto","correspondingAuthor":true,"prefix":"","firstName":"Gelareh","middleName":"","lastName":"Hajian","suffix":""},{"id":499564775,"identity":"c7e508eb-742a-4709-b114-39f868662e59","order_by":1,"name":"Bing Ye","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Bing","middleName":"","lastName":"Ye","suffix":""},{"id":499564776,"identity":"9e22ef7e-a2c6-4ae9-95d8-967c5a5db85b","order_by":2,"name":"Shabnam Haghzare","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Shabnam","middleName":"","lastName":"Haghzare","suffix":""},{"id":499564777,"identity":"1d2f4558-e851-4f09-a698-1df1aa79b514","order_by":3,"name":"Mark Rapoport","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Mark","middleName":"","lastName":"Rapoport","suffix":""},{"id":499564778,"identity":"30c54ba2-5e42-4e50-827e-1dd737cbeafa","order_by":4,"name":"Gary Naglie","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Gary","middleName":"","lastName":"Naglie","suffix":""},{"id":499564779,"identity":"9d13e50d-4e54-4401-903e-4d65f55af8f1","order_by":5,"name":"Jennifer L. Campos","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"L.","lastName":"Campos","suffix":""},{"id":499564780,"identity":"96bbe93d-8030-4794-8e39-c45c970b7a41","order_by":6,"name":"Alex Mihailidis","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Alex","middleName":"","lastName":"Mihailidis","suffix":""}],"badges":[],"createdAt":"2025-08-12 18:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7358601/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7358601/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89035721,"identity":"c6260b01-7f6d-4fe6-baab-761d3bb8b8bb","added_by":"auto","created_at":"2025-08-14 03:55:20","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":833623,"visible":true,"origin":"","legend":"","description":"","filename":"GelarehHajianAHybridUnsupervisedLearningFrameworkforSafetyAnalysis.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7358601/v1_covered_49061987-6b16-4420-9487-56e20be02cdf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Hybrid Unsupervised Learning Framework for Safety Analysis of Conditionally Automated Vehicle Use in Older Adults with Cognitive Challenges","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Automated vehicles, Older adults with cognitive impairment, Driving performance, Safety analysis, Machine Learning, Unsupervised learning","lastPublishedDoi":"10.21203/rs.3.rs-7358601/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7358601/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Driving is a critical aspect of independence and quality of life, particularly for older adults. However, cognitive challenges such as Mild Cognitive Impairment (MCI), and dementia often lead to driving cessation, resulting in negative social, emotional, and health outcomes. Conditionally Automated vehicles (CAVs) may offer opportunities as an assistive technology to extend driving years for older adults. Despite advancements in CAV technologies, significant gaps remain in ensuring safety during takeover requests (TORs) for individuals with cognitive challenges, as no studies have evaluated their performance using CAVs. Moreover, existing studies analyzing driving performance in CAVs largely focus on comparing isolated metrics such as reaction time (RT) or takeover quality (TOQ) metrics, lacking an integrated framework to assess overall safety, representing another critical research gap.\nIn this study, we address these gaps by developing a novel hybrid unsupervised learning framework to comprehensively analyze driving performance during TORs in older adults with cognitive challenges. Using data from 37 cognitively healthy older adults, we first apply Kmeans clustering combined with domain knowledge to cluster cognitively healthy participants into “safe” and “less safe” driving profiles, establishing a standardized safety baseline. This framework is then extended to evaluate CAV safety among older adults with cognitive challenges (Subjective Cognitive Decline (SCD), MCI, and dementia). To validate the robustness of our approach, we compare the hybrid framework against traditional clustering methods (Kmeans, hierarchical, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN)). \nResults demonstrate that the proposed hybrid framework outperforms traditional methods, providing a more comprehensive and individualized evaluation of CAV safety during TORs. This study highlights the importance of integrating domain knowledge with data-driven methods for developing a safety framework, paving the way for CAVs safety analysis to support older adults’ independence and mobility.","manuscriptTitle":"A Hybrid Unsupervised Learning Framework for Safety Analysis of Conditionally Automated Vehicle Use in Older Adults with Cognitive Challenges","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-14 03:47:15","doi":"10.21203/rs.3.rs-7358601/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-29T05:12:26+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-27T04:29:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-27T04:28:30+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Geriatrics","date":"2025-08-12T18:41:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-geriatrics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bgtc","sideBox":"Learn more about [BMC Geriatrics](http://bmcgeriatr.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bgtc/default.aspx","title":"BMC Geriatrics","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5f1d2432-615f-451a-a63e-73c8ba3a2e5a","owner":[],"postedDate":"August 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-13T09:10:12+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-14 03:47:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7358601","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7358601","identity":"rs-7358601","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}