From Clinical Criteria to AI-Based Classification of Advanced Parkinson’s Disease: A Data-Driven Approach Using Structured PPMI Data

From Clinical Criteria to AI-Based Classification of Advanced Parkinson’s Disease: A Data-Driven Approach Using Structured PPMI Data | 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 Article From Clinical Criteria to AI-Based Classification of Advanced Parkinson’s Disease: A Data-Driven Approach Using Structured PPMI Data Iñigo Gabilondo, Angela Sáenz, Sandra Seijo, Alvaro Ochoa, Unai Zalabarria, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7745559/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Advanced Parkinson’s disease (APD) involves severe motor and non-motor complications and requires early identification, yet lacks a standardized quantitative definition. This study translated expert-defined CDEPA criteria into 216 structured variables from the Parkinson’s Progression Markers Initiative (PPMI) dataset to train machine learning models for early APD classification. A 1,302 patients cohort was followed up for 13 years. A label-rescuing strategy addressed longitudinal incompleteness. Supervised models trained on baseline data predicted future APD status. Binary classifiers outperformed multiclass approaches; the best-performing model (XGBoost, Year 9) achieved AUC 0.881, balanced accuracy 0.824, and F1 score 0.819. Top predictors included genetic mutation status, age, MDS-UPDRS I–II, REM sleep behavior disorder, and tremor severity. Non-motor symptoms—especially autonomic dysfunction and sleep disturbances—were more informative than motor signs, comprising 57.1% of top features. These findings support the feasibility of early APD classification and propose a scalable, data-driven framework for APD prediction. Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Health sciences/Neurology Biological sciences/Neuroscience Advanced Parkinson’s Disease Artificial Intelligence Machine Learning PPMI Predictive Modelling CDEPA criteria Full Text Additional Declarations No competing interests reported. Supplementary Files SuppFIle.docx Cite Share Download PDF Status: Posted Version 1 posted 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-7745559","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":528069868,"identity":"32eef17e-c5db-4d53-95f4-1e37a37020f2","order_by":0,"name":"Iñigo Gabilondo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYBACNhBKKGBg4AfxQAwitRgwMEg2gLQYEG0RUKXBARCbGC18/IufPXhgYJNnfH514ocHBgzy/GIHCFgh8czcIMEgrdjsxtvNEkCHGc6cnUBIywEzoMrDidtunN0A0pJgcJugluPfwFo2zzi7+QdxWvh7ILZs4O/dRqwtPOUgvyTOuMG7zSLBQIKwX+T7j297+KPCJrG//+zmm0CGPL80AS0MEgkoDAkCykGA/wA6YxSMglEwCkYBGgAANz9D3u/G/akAAAAASUVORK5CYII=","orcid":"","institution":"Biobizkaia Health Research Institute","correspondingAuthor":true,"prefix":"","firstName":"Iñigo","middleName":"","lastName":"Gabilondo","suffix":""},{"id":528069870,"identity":"794cd7f4-60b2-4164-aebc-f8fc93210d70","order_by":1,"name":"Angela Sáenz","email":"","orcid":"","institution":"Biobizkaia Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Sáenz","suffix":""},{"id":528069871,"identity":"6ad7cd78-ae08-4d03-a6fb-3bb9850917cc","order_by":2,"name":"Sandra Seijo","email":"","orcid":"","institution":"Ibermática (Spain)","correspondingAuthor":false,"prefix":"","firstName":"Sandra","middleName":"","lastName":"Seijo","suffix":""},{"id":528069872,"identity":"a8eedbb7-9621-4100-92bb-827fbd932d9e","order_by":3,"name":"Alvaro Ochoa","email":"","orcid":"","institution":"Ibermática (Spain)","correspondingAuthor":false,"prefix":"","firstName":"Alvaro","middleName":"","lastName":"Ochoa","suffix":""},{"id":528069875,"identity":"4f1c877f-cd53-4b97-9830-fd9f0a53e760","order_by":4,"name":"Unai Zalabarria","email":"","orcid":"","institution":"Ibermática (Spain)","correspondingAuthor":false,"prefix":"","firstName":"Unai","middleName":"","lastName":"Zalabarria","suffix":""},{"id":528069877,"identity":"93f29d18-7a08-4f12-bf5b-a5e5621d703c","order_by":5,"name":"Itziar Cuenca","email":"","orcid":"","institution":"Ibermática (Spain)","correspondingAuthor":false,"prefix":"","firstName":"Itziar","middleName":"","lastName":"Cuenca","suffix":""},{"id":528069879,"identity":"1e5511b3-58dc-4333-86e8-7e8356623a8d","order_by":6,"name":"Beatriz Tijero","email":"","orcid":"","institution":"Hospital de Cruces","correspondingAuthor":false,"prefix":"","firstName":"Beatriz","middleName":"","lastName":"Tijero","suffix":""},{"id":528069880,"identity":"54f7a230-d804-4cd4-854d-25f625a59d24","order_by":7,"name":"Tamara Fernández-Valle","email":"","orcid":"","institution":"Hospital de Cruces","correspondingAuthor":false,"prefix":"","firstName":"Tamara","middleName":"","lastName":"Fernández-Valle","suffix":""},{"id":528069882,"identity":"d9b79221-bc87-4cd9-b587-e09706d9a0f6","order_by":8,"name":"Marta Ruiz","email":"","orcid":"","institution":"Hospital de Cruces","correspondingAuthor":false,"prefix":"","firstName":"Marta","middleName":"","lastName":"Ruiz","suffix":""},{"id":528069886,"identity":"7a6e99de-a92a-4df3-859f-3401e3fdcaeb","order_by":9,"name":"Marian Acera","email":"","orcid":"","institution":"Biobizkaia Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Marian","middleName":"","lastName":"Acera","suffix":""},{"id":528069888,"identity":"6bf4d614-f769-438f-af32-6eaf761abd73","order_by":10,"name":"Inês Sousa","email":"","orcid":"","institution":"Fraunhofer Portugal Research","correspondingAuthor":false,"prefix":"","firstName":"Inês","middleName":"","lastName":"Sousa","suffix":""},{"id":528069890,"identity":"ed7ef67d-8c15-4fc0-9a27-b3517556fe2d","order_by":11,"name":"Juan Carlos Gómez‑Esteban","email":"","orcid":"","institution":"Biobizkaia Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"Carlos","lastName":"Gómez‑Esteban","suffix":""},{"id":528069891,"identity":"80cf7a34-5378-40f7-ab3c-68db467f4e16","order_by":12,"name":"Rocío Del Pino","email":"","orcid":"","institution":"Biobizkaia Health Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Rocío","middleName":"Del","lastName":"Pino","suffix":""}],"badges":[],"createdAt":"2025-09-29 22:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7745559/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7745559/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93411333,"identity":"cddb7f35-ea8a-4a46-a07e-e676b259c65b","added_by":"auto","created_at":"2025-10-13 14:36:34","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":842045,"visible":true,"origin":"","legend":"","description":"","filename":"AIAPDPPMICDEPAAISym4MedIG12AGO2025REFNPJUSSSAORoNoEndtablefigs.docx","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/66836b2b44e1d5035552ee7d.docx"},{"id":93411331,"identity":"e45990af-aa58-4df9-903b-e58089050103","added_by":"auto","created_at":"2025-10-13 14:36:33","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12714,"visible":true,"origin":"","legend":"","description":"","filename":"641f27d4a6e84b009f3e345d75aec94c.json","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/76075e91eb384f12054cf2d3.json"},{"id":93410780,"identity":"ece05aa8-4dbf-452c-8f7f-7395f7335779","added_by":"auto","created_at":"2025-10-13 14:28:33","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26625,"visible":true,"origin":"","legend":"","description":"","filename":"SuppFIle.docx","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/f99f2587cedf2c654eefb443.docx"},{"id":93410787,"identity":"3439ab9a-4df7-4578-b4a2-063a4eb4f5a8","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":179852,"visible":true,"origin":"","legend":"","description":"","filename":"641f27d4a6e84b009f3e345d75aec94c1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/c0bf463ef862cc41942fe169.xml"},{"id":93412330,"identity":"77d2e775-4bec-4c37-8a3a-5a1e801aa0fc","added_by":"auto","created_at":"2025-10-13 14:44:34","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205388,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/5dc9532fcbfc7c35338e3305.jpeg"},{"id":93410782,"identity":"082f69ee-2a45-45c7-8928-792b197927ea","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":99787,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/5948c7b1c3e900d14a6492b9.png"},{"id":93411335,"identity":"006b92d0-5938-4706-ae4b-683f4a3e95fd","added_by":"auto","created_at":"2025-10-13 14:36:34","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":255292,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/e3ac8b37bf70e19c3dd862c4.jpeg"},{"id":93411336,"identity":"eadece5d-c374-4c92-b89f-651ba2fbfb7c","added_by":"auto","created_at":"2025-10-13 14:36:34","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":185524,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/e677b016384d80c61b706905.jpeg"},{"id":93411332,"identity":"06016dc0-07ed-4243-8e07-883f3e722047","added_by":"auto","created_at":"2025-10-13 14:36:34","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39980,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/5b45a34680126641681e8bb8.png"},{"id":93410784,"identity":"78170be0-be37-470d-ad6b-80e3e3ed1468","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":40126,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/b00ad82d4d233c91f5a15875.png"},{"id":93410788,"identity":"ac663550-46a2-4453-afda-ebf82000a666","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":46510,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/3a6a844fa4d4a13c1abb7559.png"},{"id":93410790,"identity":"e93847e1-c49c-4f33-9e44-ef8ce87b2725","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50604,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/ecc157ab3717f1d3855d0eed.png"},{"id":93410792,"identity":"5699ac2a-35ca-43d1-997c-dd68d84dabe4","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":180087,"visible":true,"origin":"","legend":"","description":"","filename":"641f27d4a6e84b009f3e345d75aec94c1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/e83d3ff1718cd06d54444545.xml"},{"id":93410793,"identity":"6b52bab7-7d64-47fc-aef9-79996d8061b1","added_by":"auto","created_at":"2025-10-13 14:28:34","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":190558,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/fc55c73ad8b8a714b047380a.html"},{"id":103433631,"identity":"8ee8cf09-f038-4ce0-aa3f-c3faa4082ab0","added_by":"auto","created_at":"2026-02-25 15:56:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1107050,"visible":true,"origin":"","legend":"","description":"","filename":"AIAPDPPMICDEPAAISym4MedIG12AGO2025REFNPJUSSSAORoNoEndtablefigs.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1_covered_aedd4de0-682c-49b0-97ff-84ca87402fae.pdf"},{"id":93410779,"identity":"65688ff9-1557-4ed3-b1b3-398293a5c729","added_by":"auto","created_at":"2025-10-13 14:28:33","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":26625,"visible":true,"origin":"","legend":"","description":"","filename":"SuppFIle.docx","url":"https://assets-eu.researchsquare.com/files/rs-7745559/v1/cf65e4a08f72ca9b3a90141f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Clinical Criteria to AI-Based Classification of Advanced Parkinson’s Disease: A Data-Driven Approach Using Structured PPMI Data","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Advanced Parkinson’s Disease, Artificial Intelligence, Machine Learning, PPMI, Predictive Modelling, CDEPA criteria","lastPublishedDoi":"10.21203/rs.3.rs-7745559/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7745559/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAdvanced Parkinson\u0026rsquo;s disease (APD) involves severe motor and non-motor complications and requires early identification, yet lacks a standardized quantitative definition. This study translated expert-defined CDEPA criteria into 216 structured variables from the Parkinson\u0026rsquo;s Progression Markers Initiative (PPMI) dataset to train machine learning models for early APD classification. A 1,302 patients cohort was followed up for 13 years. A label-rescuing strategy addressed longitudinal incompleteness. Supervised models trained on baseline data predicted future APD status. Binary classifiers outperformed multiclass approaches; the best-performing model (XGBoost, Year 9) achieved AUC 0.881, balanced accuracy 0.824, and F1 score 0.819. Top predictors included genetic mutation status, age, MDS-UPDRS I\u0026ndash;II, REM sleep behavior disorder, and tremor severity. Non-motor symptoms\u0026mdash;especially autonomic dysfunction and sleep disturbances\u0026mdash;were more informative than motor signs, comprising 57.1% of top features. These findings support the feasibility of early APD classification and propose a scalable, data-driven framework for APD prediction.\u003c/p\u003e","manuscriptTitle":"From Clinical Criteria to AI-Based Classification of Advanced Parkinson’s Disease: A Data-Driven Approach Using Structured PPMI Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-13 14:28:29","doi":"10.21203/rs.3.rs-7745559/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9d5d9b65-5974-43d7-9cc5-ef9ed01ba954","owner":[],"postedDate":"October 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56133725,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56133726,"name":"Health sciences/Diseases"},{"id":56133727,"name":"Health sciences/Neurology"},{"id":56133728,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-02-25T15:56:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-13 14:28:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7745559","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7745559","identity":"rs-7745559","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}