FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting

FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting | 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 FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting He Zhu, Jun Bai, Na Li, Xiaoxiao Li, Dianbo Liu, David Buckeridge, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6099872/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 17 May, 2025 Read the published version in npj Digital Medicine → Version 1 posted 11 You are reading this latest preprint version Abstract Federated learning (FL) enables collaborative analysis of decentralized medical data while preserving patient privacy. However, the covariate shift from demographic and clinical differences can reduce model generalizability. We propose FedWeight, a novel FL framework that mitigates covariate shift by reweighting patient data from the source sites using density estimators, allowing the trained model to better align with the distribution of the target site. To support unsupervised applications, we introduce FedWeight ETM, a federated embedded topic model. We evaluated FedWeight in cross-site FL on the eICU dataset and cross-dataset FL between eICU and MIMIC III. FedWeight consistently outperforms standard FL baselines in predicting ICU mortality, ventilator use, sepsis diagnosis, and length of stay. SHAP-based interpretation and ETM-based topic modeling reveal improved identification of clinically relevant characteristics and disease topics associated with ICU readmission. Health sciences/Health care/Diagnosis Health sciences/Health care/Public health/Epidemiology Full Text Additional Declarations No competing interests reported. Supplementary Files FedWeightrevisionsuppl.pdf Cite Share Download PDF Status: Published Journal Publication published 17 May, 2025 Read the published version in npj Digital Medicine → Version 1 posted Editorial decision: Accepted 21 Apr, 2025 Reviews received at journal 21 Apr, 2025 Reviews received at journal 19 Apr, 2025 Reviews received at journal 13 Apr, 2025 Reviewers agreed at journal 13 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers invited by journal 11 Apr, 2025 Submission checks completed at journal 11 Apr, 2025 First submitted to journal 31 Mar, 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-6099872","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":442284581,"identity":"c0960107-94e9-4bd5-906b-998d21c1a551","order_by":0,"name":"He Zhu","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"He","middleName":"","lastName":"Zhu","suffix":""},{"id":442284582,"identity":"bdd4bdcf-88cb-4f65-8299-82752d304883","order_by":1,"name":"Jun Bai","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Bai","suffix":""},{"id":442284583,"identity":"3ca06870-da15-48be-8bb5-6b3e16ad6fea","order_by":2,"name":"Na Li","email":"","orcid":"","institution":"Department of Community Health Sciences, University of Calgary","correspondingAuthor":false,"prefix":"","firstName":"Na","middleName":"","lastName":"Li","suffix":""},{"id":442284584,"identity":"719c36a9-06b5-40c0-98d4-60c7cd5635b9","order_by":3,"name":"Xiaoxiao Li","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Xiaoxiao","middleName":"","lastName":"Li","suffix":""},{"id":442284585,"identity":"6e83a6fb-cac7-4041-a8dc-3910e03dfa3d","order_by":4,"name":"Dianbo Liu","email":"","orcid":"","institution":"National University of Singapore","correspondingAuthor":false,"prefix":"","firstName":"Dianbo","middleName":"","lastName":"Liu","suffix":""},{"id":442284586,"identity":"329d79f4-64ca-4e36-9d89-f85dc5c44b93","order_by":5,"name":"David Buckeridge","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Buckeridge","suffix":""},{"id":442284587,"identity":"950e77ca-e045-417e-8863-d1fbab71ea5c","order_by":6,"name":"Yue Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0ElEQVRIiWNgGAWjYHACxgMJFVBmQoEFcXoOJJyBaTGQIFILYxuMSYwWgxvJBw48nGdnzyB2+PGLB0At/O0HCGlJSziQuC05sUE6zcwC5DCJMwmEtOQYALUcSGCQzmEzAGkxYCCoJf/DgcQ5B+wRWvgfELSF4UBiwwHGBukc5gdgLRIEbJE888zgQMKx5MQ2oF9AgcwjcYOALXzHkx8+/FFjZ88vnfz4448KGzn+fgK2KByAMtiACBQpPPjVA4F8A4LN/IGg8lEwCkbBKBiRAAANIkL1pnQaZgAAAABJRU5ErkJggg==","orcid":"","institution":"McGill University","correspondingAuthor":true,"prefix":"","firstName":"Yue","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-02-24 21:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6099872/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6099872/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41746-025-01661-8","type":"published","date":"2025-05-17T15:56:51+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83067648,"identity":"fb1e9506-dde5-433f-9bda-d51b18f383a9","added_by":"auto","created_at":"2025-05-19 16:01:24","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7458468,"visible":true,"origin":"","legend":"","description":"","filename":"FedWeightrevisionlatexsource.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6099872/v1_covered_fc056133-97f0-46e4-9e49-172b50090887.pdf"},{"id":80526250,"identity":"207d8383-38bb-442e-9a2a-2fdc1cc9a360","added_by":"auto","created_at":"2025-04-14 10:01:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2377393,"visible":true,"origin":"","legend":"","description":"","filename":"FedWeightrevisionsuppl.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6099872/v1/3320b78a5ecdb401d7afef7f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6099872/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6099872/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Federated learning (FL) enables collaborative analysis of decentralized medical data while preserving patient privacy. However, the covariate shift from demographic and clinical differences can reduce model generalizability. We propose FedWeight, a novel FL framework that mitigates covariate shift by reweighting patient data from the source sites using density estimators, allowing the trained model to better align with the distribution of the target site. To support unsupervised applications, we introduce FedWeight ETM, a federated embedded topic model. We evaluated FedWeight in cross-site FL on the eICU dataset and cross-dataset FL between eICU and MIMIC III. FedWeight consistently outperforms standard FL baselines in predicting ICU mortality, ventilator use, sepsis diagnosis, and length of stay. SHAP-based interpretation and ETM-based topic modeling reveal improved identification of clinically relevant characteristics and disease topics associated with ICU readmission.","manuscriptTitle":"FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-14 10:01:19","doi":"10.21203/rs.3.rs-6099872/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Accepted","date":"2025-04-21T23:42:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-21T14:30:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-19T10:15:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-14T03:19:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"325512052260736610141794729917577471567","date":"2025-04-13T14:25:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"111780100576070012238413825806638855435","date":"2025-04-11T12:10:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"261872873084191358998642731614806185952","date":"2025-04-11T11:16:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"17903361789280087717381572893422757483","date":"2025-04-11T07:38:00+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-11T05:30:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-11T05:13:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-04-01T00:09:59+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-digital-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjdigitalmed","sideBox":"Learn more about [npj Digital Medicine](http://www.nature.com/npjdigitalmed/)","snPcode":"41746","submissionUrl":"https://submission.springernature.com/new-submission/41746/3","title":"npj Digital Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dec59e95-dd19-469c-b43e-fdadf91b0175","owner":[],"postedDate":"April 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":47087677,"name":"Health sciences/Health care/Diagnosis"},{"id":47087678,"name":"Health sciences/Health care/Public health/Epidemiology"}],"tags":[],"updatedAt":"2025-05-19T15:58:31+00:00","versionOfRecord":{"articleIdentity":"rs-6099872","link":"https://doi.org/10.1038/s41746-025-01661-8","journal":{"identity":"npj-digital-medicine","isVorOnly":false,"title":"npj Digital Medicine"},"publishedOn":"2025-05-17 15:56:51","publishedOnDateReadable":"May 17th, 2025"},"versionCreatedAt":"2025-04-14 10:01:19","video":"","vorDoi":"10.1038/s41746-025-01661-8","vorDoiUrl":"https://doi.org/10.1038/s41746-025-01661-8","workflowStages":[]},"version":"v1","identity":"rs-6099872","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6099872","identity":"rs-6099872","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}