Enhancing Classification of rare white blood cells in FPM with a Physics-inspired GAN

Enhancing Classification of rare white blood cells in FPM with a Physics-inspired GAN | 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 Enhancing Classification of rare white blood cells in FPM with a Physics-inspired GAN Houda Hassini, Bernadette Dorizzi, Vincent Leymarie, Jacques Klossa, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7262018/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 27 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted 15 You are reading this latest preprint version Abstract In this work, we propose a novel GAN-based architecture, termed Physics-Inspired GAN (PI-GAN), to generate synthetic bimodal data comprising both intensity and phase images as produced through Fourier Ptychographic Microscopy (FPM). By explicitly incorporating the forward model of image formation into the GAN architecture, our approach ensures that the physical relationship between the intensity and phase modalities is preserved throughout the training and generation processes, therefore solving the mode collapse problem encountered in classical GANs. Our approach is evaluated for the classification of the five major types of white blood cells (WBCs) in peripheral blood smears, a domain where severe class imbalance is a major challenge. In particular, basophils represent less than 1% of circulating WBCs, making it difficult to train robust classifiers without synthetic augmentation. To overcome the scarcity of basophil data, we proposed a two-step fine-tuning strategy: first training the PI-GAN to generate neutrophils (a more abundant but morphologically similar class), and then adapting the model to produce basophils. Our results show that the addition of synthetic basophil images allows a great improvement (5% in precision) in the ability to correctly classify basophils. Our approach offers great potential for future hybrid models that combine physics-based priors with the flexibility of deep generative networks. Biological sciences/Computational biology and bioinformatics Physical sciences/Engineering Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 27 Nov, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 08 Oct, 2025 Reviews received at journal 26 Sep, 2025 Reviews received at journal 25 Sep, 2025 Reviews received at journal 23 Sep, 2025 Reviews received at journal 19 Sep, 2025 Reviewers agreed at journal 05 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers invited by journal 03 Sep, 2025 Editor invited by journal 28 Aug, 2025 Editor assigned by journal 22 Aug, 2025 Submission checks completed at journal 19 Aug, 2025 First submitted to journal 19 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-7262018","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":510830200,"identity":"fc4c8ef4-665e-4d3a-bcf6-dd70a848305a","order_by":0,"name":"Houda Hassini","email":"","orcid":"","institution":"Samovar, T ´ eí ecom SudParis, Institut Polytechnique de Paris","correspondingAuthor":false,"prefix":"","firstName":"Houda","middleName":"","lastName":"Hassini","suffix":""},{"id":510830201,"identity":"d5410dc1-119c-4ffd-ac04-91309027bef4","order_by":1,"name":"Bernadette Dorizzi","email":"","orcid":"","institution":"Samovar, T ´ eí ecom SudParis, Institut Polytechnique de Paris","correspondingAuthor":false,"prefix":"","firstName":"Bernadette","middleName":"","lastName":"Dorizzi","suffix":""},{"id":510830203,"identity":"9425066f-35d8-45c9-b843-ac5895efd927","order_by":2,"name":"Vincent Leymarie","email":"","orcid":"","institution":"HUPSSD, H ˆ opital Avicenne, AP-HP","correspondingAuthor":false,"prefix":"","firstName":"Vincent","middleName":"","lastName":"Leymarie","suffix":""},{"id":510830204,"identity":"5cf16947-c946-4d1b-906c-36b18d4fb3c8","order_by":3,"name":"Jacques Klossa","email":"","orcid":"","institution":"TRIBVN","correspondingAuthor":false,"prefix":"","firstName":"Jacques","middleName":"","lastName":"Klossa","suffix":""},{"id":510830208,"identity":"7ff547bf-d552-449a-9bd0-0bd9fa59e5f8","order_by":4,"name":"Yaneck Gottesman","email":"data:image/png;base64,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","orcid":"","institution":"Samovar, T ´ eí ecom SudParis, Institut Polytechnique de Paris","correspondingAuthor":true,"prefix":"","firstName":"Yaneck","middleName":"","lastName":"Gottesman","suffix":""}],"badges":[],"createdAt":"2025-07-31 12:08:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7262018/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7262018/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-26429-8","type":"published","date":"2025-11-27T15:58:21+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97178591,"identity":"cb6ff7d4-7731-4b99-9aed-6c2af676b9b2","added_by":"auto","created_at":"2025-12-01 16:11:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":935697,"visible":true,"origin":"","legend":"","description":"","filename":"EnhancingClassificationofrarewhitebloodcellsinFPMwithaPhysicsinspiredGANR2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7262018/v1_covered_6b8ac733-13f1-4249-ada0-b8aad61797d4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Enhancing Classification of rare white blood cells in FPM with a Physics-inspired GAN","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7262018/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7262018/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"In this work, we propose a novel GAN-based architecture, termed Physics-Inspired GAN (PI-GAN), to generate synthetic bimodal data comprising both intensity and phase images as produced through Fourier Ptychographic Microscopy (FPM). By explicitly incorporating the forward model of image formation into the GAN architecture, our approach ensures that the physical relationship between the intensity and phase modalities is preserved throughout the training and generation processes, therefore solving the mode collapse problem encountered in classical GANs. Our approach is evaluated for the classification of the five major types of white blood cells (WBCs) in peripheral blood smears, a domain where severe class imbalance is a major challenge. In particular, basophils represent less than 1% of circulating WBCs, making it difficult to train robust classifiers without synthetic augmentation. To overcome the scarcity of basophil data, we proposed a two-step fine-tuning strategy: first training the PI-GAN to generate neutrophils (a more abundant but morphologically similar class), and then adapting the model to produce basophils. Our results show that the addition of synthetic basophil images allows a great improvement (5% in precision) in the ability to correctly classify basophils. Our approach offers great potential for future hybrid models that combine physics-based priors with the flexibility of deep generative networks.","manuscriptTitle":"Enhancing Classification of rare white blood cells in FPM with a Physics-inspired GAN","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-10 04:19:15","doi":"10.21203/rs.3.rs-7262018/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-08T08:21:44+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-26T04:02:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-26T01:52:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-23T22:17:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-19T23:46:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6375389454646497489733811507356757579","date":"2025-09-05T13:49:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"238602292818491148104839459740114955160","date":"2025-09-03T20:56:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"130388215923450772590392321320842464464","date":"2025-09-03T15:58:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"290964699060060293657368277005955726565","date":"2025-09-03T14:20:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"336445115785728093386104025181173684618","date":"2025-09-03T14:19:53+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-03T13:43:37+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-08-28T07:58:48+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-22T17:03:24+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-19T11:52:41+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-08-19T11:51:22+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4e744746-031c-43cf-8885-4e357ac629e0","owner":[],"postedDate":"September 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":54264856,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":54264857,"name":"Physical sciences/Engineering"},{"id":54264858,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-12-01T16:04:13+00:00","versionOfRecord":{"articleIdentity":"rs-7262018","link":"https://doi.org/10.1038/s41598-025-26429-8","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-11-27 15:58:21","publishedOnDateReadable":"November 27th, 2025"},"versionCreatedAt":"2025-09-10 04:19:15","video":"","vorDoi":"10.1038/s41598-025-26429-8","vorDoiUrl":"https://doi.org/10.1038/s41598-025-26429-8","workflowStages":[]},"version":"v1","identity":"rs-7262018","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7262018","identity":"rs-7262018","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}