Target Informed Client Recruitment for Efficient Federated Learning in Healthcare

Target Informed Client Recruitment for Efficient Federated Learning in Healthcare | 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 Target Informed Client Recruitment for Efficient Federated Learning in Healthcare Vincent Scheltjens, Lyse Naomi Wamba Momo, Wouter Verbeke, Bart De Moor This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3825499/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Dec, 2024 Read the published version in BMC Medical Informatics and Decision Making → Version 1 posted 10 You are reading this latest preprint version Abstract Background: Modern machine learning and deep learning methods have been widely incorporated in decision making processes in healthcare in the form of decision support mechanisms. In healthcare, data are abundant but typically not centrally available and, therefore, require some form of aggregation to facilitate training procedures. Aggregating sensitive data poses a significant privacy risk, which is why, both in Europe and the United States, legal frameworks regulate the treatment of such data. Whilst these measures protect the individual behind the data, they pose a significant challenge that results in extensive legal administration related to data sharing efforts. Federated learning (FL) offers a way to mitigate these challenges by allowing to learn models in distributed fashion, eliminating the need to aggregate data for the purpose of training. However, FL comes with a new set of challenges related to communication overhead, client selection and efficiency of the FL training procedure, among others. Methods: In this work, we extend on a previously proposed client recruitment approach by incorporating knowledge on the local hardware such that it becomes possible to recruit a subset of clients for the federation based on the construct of client-level representativeness, which is expressed in terms of the local target distribution divergence, sample size, and the underlying hardware. Results: We show that, for prominent, medical regression and classification tasks, the recruitment approach yields results that are on par, or better, compared to the central and federated approaches. The proposed approach requires a mere fraction of the data for training and reduces the training time by a factor of 3-4. In addition, we show that excluded clients can still significantly benefit from the resulting federated model through local fine-tuning. Conclusions: By expressing the representativeness of clients in function of the deviation in the local target distribution, the sample size and efficiency of the underlying hardware, we are able to define a recruitment approach that yields a subset of clients for the federation resulting in significantly reduced training time, without harming predictive performance, whilst improving the privacy preserving characteristics compared to the standard FL and central approaches. Federated Learning Client Recruitment Deep Learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 18 Dec, 2024 Read the published version in BMC Medical Informatics and Decision Making → Version 1 posted Editorial decision: Revision requested 18 Jun, 2024 Reviews received at journal 04 Jun, 2024 Reviewers agreed at journal 20 May, 2024 Reviews received at journal 28 Mar, 2024 Reviewers agreed at journal 18 Mar, 2024 Reviewers agreed at journal 28 Feb, 2024 Reviewers invited by journal 30 Jan, 2024 Editor assigned by journal 03 Jan, 2024 Submission checks completed at journal 03 Jan, 2024 First submitted to journal 31 Dec, 2023 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-3825499","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":264958185,"identity":"22d6efcd-e5fd-4412-a60f-f2885d0a2056","order_by":0,"name":"Vincent Scheltjens","email":"data:image/png;base64,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","orcid":"","institution":"KU Leuven","correspondingAuthor":true,"prefix":"","firstName":"Vincent","middleName":"","lastName":"Scheltjens","suffix":""},{"id":264958186,"identity":"77244717-c5f9-4f44-91d8-b79ff35fb5ba","order_by":1,"name":"Lyse Naomi Wamba Momo","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Lyse","middleName":"Naomi Wamba","lastName":"Momo","suffix":""},{"id":264958187,"identity":"ebca6e27-2161-43fb-b901-9d203dc576ee","order_by":2,"name":"Wouter Verbeke","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Wouter","middleName":"","lastName":"Verbeke","suffix":""},{"id":264958188,"identity":"d8524684-2fd9-4c04-be94-df17069dd418","order_by":3,"name":"Bart De Moor","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Bart","middleName":"","lastName":"De Moor","suffix":""}],"badges":[],"createdAt":"2023-12-31 08:59:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3825499/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3825499/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12911-024-02798-4","type":"published","date":"2024-12-18T15:58:15+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":72201977,"identity":"814daeb7-9d1c-4d8b-80c9-2b287be0eaee","added_by":"auto","created_at":"2024-12-23 16:13:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":928028,"visible":true,"origin":"","legend":"","description":"","filename":"BMCCLRCFINAL.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3825499/v1_covered_2f6a25f6-2ee2-47f3-bf50-395f5dc82f40.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Target Informed Client Recruitment for Efficient Federated Learning in Healthcare","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":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Federated Learning, Client Recruitment, Deep Learning","lastPublishedDoi":"10.21203/rs.3.rs-3825499/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3825499/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Modern machine learning and deep learning methods have been widely incorporated in decision making processes in healthcare in the form of decision support mechanisms. In healthcare, data are abundant but typically not centrally available and, therefore, require some form of aggregation to facilitate training procedures. Aggregating sensitive data poses a significant privacy risk, which is why, both in Europe and the United States, legal frameworks regulate the treatment of such data. Whilst these measures protect the individual behind the data, they pose a significant challenge that results in extensive legal administration related to data sharing efforts. Federated learning (FL) offers a way to mitigate these challenges by allowing to learn models in distributed fashion, eliminating the need to aggregate data for the purpose of training. However, FL comes with a new set of challenges related to communication overhead, client selection and efficiency of the FL training procedure, among others.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e In this work, we extend on a previously proposed client recruitment approach by incorporating knowledge on the local hardware such that it becomes possible to recruit a subset of clients for the federation based on the construct of client-level representativeness, which is expressed in terms of the local target distribution divergence, sample size, and the underlying hardware.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e We show that, for prominent, medical regression and classification tasks, the recruitment approach yields results that are on par, or better, compared to the central and federated approaches. The proposed approach requires a mere fraction of the data for training and reduces the training time by a factor of 3-4. In addition, we show that excluded clients can still significantly benefit from the resulting federated model through local fine-tuning.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e\u0026nbsp;By expressing the representativeness of clients in function of the deviation in the local target distribution, the sample size and efficiency of the underlying hardware, we are able to define a recruitment approach that yields a subset of clients for the federation resulting in significantly reduced training time, without harming predictive performance, whilst improving the privacy preserving characteristics compared to the standard FL and central approaches.\u003c/p\u003e","manuscriptTitle":"Target Informed Client Recruitment for Efficient Federated Learning in Healthcare","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-04 08:42:19","doi":"10.21203/rs.3.rs-3825499/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-06-18T12:49:12+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-06-04T06:45:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"79032732700968125829741428479205108724","date":"2024-05-20T06:08:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-28T23:00:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"2745ae01-3e2f-4375-b9ac-4581efb7ab27","date":"2024-03-18T23:25:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"4ea5cf64-419c-4a39-a2ff-36408ee080ec","date":"2024-02-28T10:40:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-01-30T23:34:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-01-03T07:30:23+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-01-03T07:26:36+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medical Informatics and Decision Making","date":"2023-12-31T08:52:23+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medical-informatics-and-decision-making","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"midm","sideBox":"Learn more about [BMC Medical Informatics and Decision Making](http://bmcmedinformdecismak.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/midm/default.aspx","title":"BMC Medical Informatics and Decision Making","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"35aed04e-91b8-44ef-8ef5-c6c18f8934e9","owner":[],"postedDate":"January 4th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-12-23T16:04:58+00:00","versionOfRecord":{"articleIdentity":"rs-3825499","link":"https://doi.org/10.1186/s12911-024-02798-4","journal":{"identity":"bmc-medical-informatics-and-decision-making","isVorOnly":false,"title":"BMC Medical Informatics and Decision Making"},"publishedOn":"2024-12-18 15:58:15","publishedOnDateReadable":"December 18th, 2024"},"versionCreatedAt":"2024-01-04 08:42:19","video":"","vorDoi":"10.1186/s12911-024-02798-4","vorDoiUrl":"https://doi.org/10.1186/s12911-024-02798-4","workflowStages":[]},"version":"v1","identity":"rs-3825499","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3825499","identity":"rs-3825499","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}