Flexible and Scalable Federated Learning with Deep Feature Prompts for Digital Pathology

Flexible and Scalable Federated Learning with Deep Feature Prompts for Digital Pathology | 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 Flexible and Scalable Federated Learning with Deep Feature Prompts for Digital Pathology Cong Cong, Yang Song, Antonio Di Ieva, Angela Chou, Anthony J Gill, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7060589/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Collaborative learning across medical institutions is considered essential for developing robust and generalisable models in digital pathology. Federated learning (FL) offers a promising framework for enabling collaboration without the need to centralise data. Still, its practical adoption remains limited due to challenges such as high communication overhead, model heterogeneity across institutions, and privacy concerns. To address these challenges, we propose Federated Deep Feature Prompting (FedDFP), a novel and efficient FL paradigm designed to operate in real-world, heterogeneous clinical environments. FedDFP introduces a lightweight, learnable, client-specific prompt applied to patch-level embeddings extracted from whole slide images. By sharing only these compact prompts across clients ( i.e. , participating centres), FedDFP drastically reduces communication costs by over 99.9% compared to standard FL frameworks, with improved classification accuracy across clients. Through comprehensive evaluations on three public datasets (TCGA-IDH, CAMELYON16, and CAMELYON17), which include breast cancer and glioma classification tasks, FedDFP consistently outperforms both standard and personalised FL baselines. It maintains strong performance across diverse feature extractors and heterogeneous MIL classifiers, demonstrating flexibility, scalability, and robustness. These results highlight FedDFP as a promising solution for efficient, accurate, and privacy-conscious federated learning in digital pathology. Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Health sciences/Health care Physical sciences/Mathematics and computing WSI Classification Federated Learning Visual Prompt Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 08 Dec, 2025 Reviews received at journal 05 Dec, 2025 Reviewers agreed at journal 28 Nov, 2025 Reviews received at journal 29 Aug, 2025 Reviewers agreed at journal 15 Aug, 2025 Reviewers agreed at journal 04 Aug, 2025 Reviewers invited by journal 02 Aug, 2025 Editor assigned by journal 15 Jul, 2025 Submission checks completed at journal 15 Jul, 2025 First submitted to journal 06 Jul, 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-7060589","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":495615089,"identity":"340f96e2-b344-48e6-a3ea-434a18eefc8c","order_by":0,"name":"Cong Cong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYDACCQST8QGYOkBABw+SFmYDkrWwSRClxV66+eiGH38Y5Pml269V3ShjkOO7kcD4mQefLTLH0m72tjEYzpxzpux2zjkGY8kbCczSeLVI5Jjd4G1gSDC4kZN2O7eNIXHDjQQGglpu/vkD0VIM1FIP1ML8m5CW2zxsIC3px5iBWoCMBDb8ttxIS7st2yZhOHNGDrN0zjkg48zDNss5eLSwz0g+dvPNHxt5fon0h59zymzk+Y4nH77xBo8WKADFCI8BNGoYGwhrgFr4AKiFWMWjYBSMglEwkgAAfVNKjF3CVicAAAAASUVORK5CYII=","orcid":"","institution":"Macquarie University","correspondingAuthor":true,"prefix":"","firstName":"Cong","middleName":"","lastName":"Cong","suffix":""},{"id":495615090,"identity":"59efed33-41ad-4b3b-a8ab-3c8a96e27e87","order_by":1,"name":"Yang Song","email":"","orcid":"","institution":"UNSW Sydney","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Song","suffix":""},{"id":495615091,"identity":"a16c28d6-58c1-465f-bb1f-1b9e7ed6f316","order_by":2,"name":"Antonio Di Ieva","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Antonio","middleName":"Di","lastName":"Ieva","suffix":""},{"id":495615092,"identity":"067b65a1-4911-4ae1-a924-4dfb2d30d06f","order_by":3,"name":"Angela Chou","email":"","orcid":"","institution":"NSW Health Pathology","correspondingAuthor":false,"prefix":"","firstName":"Angela","middleName":"","lastName":"Chou","suffix":""},{"id":495615093,"identity":"b804d290-6239-4219-9250-26f00af1c913","order_by":4,"name":"Anthony J Gill","email":"","orcid":"","institution":"NSW Health Pathology","correspondingAuthor":false,"prefix":"","firstName":"Anthony","middleName":"J","lastName":"Gill","suffix":""},{"id":495615094,"identity":"5dd5cf6d-0a8a-4912-aa59-c219690f452d","order_by":5,"name":"Enrico Coiera","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Enrico","middleName":"","lastName":"Coiera","suffix":""},{"id":495615095,"identity":"9c3815a5-a9a7-46e6-829f-a8df75d1ae78","order_by":6,"name":"Sidong Liu","email":"","orcid":"","institution":"Macquarie University","correspondingAuthor":false,"prefix":"","firstName":"Sidong","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-07-07 02:38:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7060589/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7060589/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88595219,"identity":"3fe7c8b1-b0df-42f2-82e4-226776fd0093","added_by":"auto","created_at":"2025-08-08 06:45:23","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2638469,"visible":true,"origin":"","legend":"","description":"","filename":"npjdmFedDFP.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7060589/v1_covered_067ae7cc-9fc8-4a55-9393-efe594edb447.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Flexible and Scalable Federated Learning with Deep Feature Prompts for Digital Pathology","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":"WSI Classification, Federated Learning, Visual Prompt","lastPublishedDoi":"10.21203/rs.3.rs-7060589/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7060589/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCollaborative learning across medical institutions is considered essential for developing robust and generalisable models in digital pathology. Federated learning (FL) offers a promising framework for enabling collaboration without the need to centralise data. Still, its practical adoption remains limited due to challenges such as high communication overhead, model heterogeneity across institutions, and privacy concerns. To address these challenges, we propose Federated Deep Feature Prompting (FedDFP), a novel and efficient FL paradigm designed to operate in real-world, heterogeneous clinical environments. FedDFP introduces a lightweight, learnable, client-specific prompt applied to patch-level embeddings extracted from whole slide images. By sharing only these compact prompts across clients (\u003cem\u003ei.e.\u003c/em\u003e, participating centres), FedDFP drastically reduces communication costs by over 99.9% compared to standard FL frameworks, with improved classification accuracy across clients. Through comprehensive evaluations on three public datasets (TCGA-IDH, CAMELYON16, and CAMELYON17), which include breast cancer and glioma classification tasks, FedDFP consistently outperforms both standard and personalised FL baselines. It maintains strong performance across diverse feature extractors and heterogeneous MIL classifiers, demonstrating flexibility, scalability, and robustness. These results highlight FedDFP as a promising solution for efficient, accurate, and privacy-conscious federated learning in digital pathology.\u003c/p\u003e","manuscriptTitle":"Flexible and Scalable Federated Learning with Deep Feature Prompts for Digital Pathology","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-08 06:29:17","doi":"10.21203/rs.3.rs-7060589/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-08T05:20:56+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-05T18:09:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"214722667168755402444514890048128625541","date":"2025-11-28T12:30:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-30T00:11:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"337895706370649246890852170467588635800","date":"2025-08-15T18:14:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"14066361793111896362542589366958670355","date":"2025-08-04T20:14:02+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-02T19:58:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-16T00:20:03+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-15T04:04:50+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Digital Medicine","date":"2025-07-07T02:35:19+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":"d3f34146-cf02-4627-968d-2d692c58cc07","owner":[],"postedDate":"August 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":52636038,"name":"Biological sciences/Cancer"},{"id":52636039,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":52636040,"name":"Health sciences/Health care"},{"id":52636041,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-04-25T09:09:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-08 06:29:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7060589","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7060589","identity":"rs-7060589","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}