A histomorphological atlas of resected mesothelioma discovered by self-supervised learning from 3446 whole-slide images

A histomorphological atlas of resected mesothelioma discovered by self-supervised learning from 3446 whole-slide images | 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 A histomorphological atlas of resected mesothelioma discovered by self-supervised learning from 3446 whole-slide images John Le Quesne, Farzaneh Seyedshahi, Kai Rakovic, Nicolas Poulain, and 14 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5678715/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 Oct, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Mesothelioma is a highly lethal and poorly biologically understood disease which presents diagnostic challenges due to its morphological complexity. This study uses self-supervised AI (Artificial Intelligence) to map the histomorphological landscape of the disease. The resulting atlas consists of recurrent patterns identified from 3446 Hematoxylin and Eosin (H&E) stained images scanned from resected tumour slides. These patterns generate highly interpretable predictions, achieving state-of-the-art performance with 0.65 concordance index (c-index) for outcomes and 85% AUC in subtyping. Their clinical relevance is endorsed by comprehensive human pathological assessment. Furthermore, we characterise the molecular underpinnings of these diverse, meaningful, predictive patterns. Our approach both improves diagnosis and deepens our understanding of mesothelioma biology, highlighting the power of this self-learning method in clinical applications and scientific discovery. Biological sciences/Cancer/Mesothelioma Biological sciences/Computational biology and bioinformatics/Machine learning Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SUPHPLMeso.pdf Supplementary Info Cite Share Download PDF Status: Published Journal Publication published 07 Oct, 2025 Read the published version in Nature Communications → 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-5678715","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":396899352,"identity":"6e32d259-99e1-4adf-acaf-5257329c22ab","order_by":0,"name":"John Le Quesne","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-3552-7446","institution":"Cancer Research UK Scotland Institute","correspondingAuthor":true,"prefix":"","firstName":"John","middleName":"Le","lastName":"Quesne","suffix":""},{"id":396899353,"identity":"578e0894-c8f8-4634-b308-955b8be990f8","order_by":1,"name":"Farzaneh Seyedshahi","email":"","orcid":"https://orcid.org/0009-0007-1111-052X","institution":"Cancer Research UK Scotland Institute","correspondingAuthor":false,"prefix":"","firstName":"Farzaneh","middleName":"","lastName":"Seyedshahi","suffix":""},{"id":396899354,"identity":"8183415b-eb52-47aa-b6cd-92c2fe47b4b2","order_by":2,"name":"Kai Rakovic","email":"","orcid":"","institution":"School of Cancer Sciences, University of Glasgow","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Rakovic","suffix":""},{"id":396899355,"identity":"87dca1e8-aac2-4120-82db-68093ef51489","order_by":3,"name":"Nicolas Poulain","email":"","orcid":"","institution":"Cancer Research UK Scotland Institute","correspondingAuthor":false,"prefix":"","firstName":"Nicolas","middleName":"","lastName":"Poulain","suffix":""},{"id":396899356,"identity":"61b3d807-4a41-43da-aa31-7063784bdfcf","order_by":4,"name":"Adalberto Claudio Quiros","email":"","orcid":"","institution":"University of Glasgow","correspondingAuthor":false,"prefix":"","firstName":"Adalberto","middleName":"Claudio","lastName":"Quiros","suffix":""},{"id":396899357,"identity":"dc9aa06c-3120-4aef-9f55-315f2b63ac07","order_by":5,"name":"Ian Powley","email":"","orcid":"https://orcid.org/0009-0009-6079-6673","institution":"CRUK Scotland Institute","correspondingAuthor":false,"prefix":"","firstName":"Ian","middleName":"","lastName":"Powley","suffix":""},{"id":396899358,"identity":"14ff3a6a-e678-4ec6-82e3-c578db6a4edd","order_by":6,"name":"Cathy Richards","email":"","orcid":"","institution":"Leicester Royal Infirmary","correspondingAuthor":false,"prefix":"","firstName":"Cathy","middleName":"","lastName":"Richards","suffix":""},{"id":396899359,"identity":"d6bbe6ce-5606-4437-8136-89ff964d8f94","order_by":7,"name":"Hussein Uraiby","email":"","orcid":"","institution":"University Hospitals of Leicester","correspondingAuthor":false,"prefix":"","firstName":"Hussein","middleName":"","lastName":"Uraiby","suffix":""},{"id":396899360,"identity":"2a5c209a-90d1-4cad-8593-a885873e5233","order_by":8,"name":"Sonja Klebe","email":"","orcid":"","institution":"Flinders Health and Medical Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Sonja","middleName":"","lastName":"Klebe","suffix":""},{"id":396899361,"identity":"219fa01a-8faf-4cd1-ae1d-71143fe821f4","order_by":9,"name":"Apostolos Nakas","email":"","orcid":"","institution":"University of Leicester \u0026 University Hospitals of Leicester NHS Trust","correspondingAuthor":false,"prefix":"","firstName":"Apostolos","middleName":"","lastName":"Nakas","suffix":""},{"id":396899362,"identity":"2ffd9a50-7f93-4f84-adaf-c38c7104d1ff","order_by":10,"name":"Claire Wilson","email":"","orcid":"https://orcid.org/0000-0002-6401-2014","institution":"Leicester Medical School, University of Leicester","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"","lastName":"Wilson","suffix":""},{"id":396899363,"identity":"186e5615-d7fd-4bd5-ae1f-7eef56d7064e","order_by":11,"name":"Marco Sereno","email":"","orcid":"https://orcid.org/0000-0003-4573-9303","institution":"University of Leicester","correspondingAuthor":false,"prefix":"","firstName":"Marco","middleName":"","lastName":"Sereno","suffix":""},{"id":396899364,"identity":"b9d90b17-a38c-44bd-81fe-b2cfb83e53ff","order_by":12,"name":"Leah Officer-Jones","email":"","orcid":"https://orcid.org/0000-0002-3690-2386","institution":"Cancer Research UK Scotland Instutute","correspondingAuthor":false,"prefix":"","firstName":"Leah","middleName":"","lastName":"Officer-Jones","suffix":""},{"id":396899365,"identity":"36f23a19-647e-421a-954d-30e81e0cf792","order_by":13,"name":"Catherine Ficken","email":"","orcid":"","institution":"Cancer Research UK Scotland Institute","correspondingAuthor":false,"prefix":"","firstName":"Catherine","middleName":"","lastName":"Ficken","suffix":""},{"id":396899366,"identity":"6cb22b19-2b39-4056-8090-c9299c3e34f7","order_by":14,"name":"Ana Teodosio","email":"","orcid":"","institution":"Birmingham Tissue Analytics, University of Birmingham","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Teodosio","suffix":""},{"id":396899367,"identity":"2f17406f-fda2-4457-9213-0275afbd6d4a","order_by":15,"name":"Fiona Ballantyne","email":"","orcid":"","institution":"Cancer Research UK Scotland Instutute","correspondingAuthor":false,"prefix":"","firstName":"Fiona","middleName":"","lastName":"Ballantyne","suffix":""},{"id":396899368,"identity":"15e69ad9-c175-4ab9-865b-b7866fdfdc0f","order_by":16,"name":"Daniel Murphy","email":"","orcid":"https://orcid.org/0000-0002-5538-5468","institution":"University of Glasgow","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Murphy","suffix":""},{"id":396899369,"identity":"b4a04157-6e33-491b-8155-a320651aabc0","order_by":17,"name":"Ke Yuan","email":"","orcid":"https://orcid.org/0000-0002-2318-1460","institution":"University of Glasgow","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Yuan","suffix":""}],"badges":[],"createdAt":"2024-12-19 17:25:57","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5678715/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5678715/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-63846-9","type":"published","date":"2025-10-07T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93008933,"identity":"437e899e-4dcf-45ea-a619-0ae4747bc9ad","added_by":"auto","created_at":"2025-10-08 07:05:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4995602,"visible":true,"origin":"","legend":"Article File","description":"","filename":"HPLMesoL.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5678715/v1_covered_26c0fc8f-007d-496c-b0ce-3ae5ad7c9f7c.pdf"},{"id":74016266,"identity":"0d709640-eb9f-4cb4-8972-2cbd795b1813","added_by":"auto","created_at":"2025-01-17 04:02:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":29776208,"visible":true,"origin":"","legend":"Supplementary Info","description":"","filename":"SUPHPLMeso.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5678715/v1/2a1ac06a0e68be92bf074ce0.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"A histomorphological atlas of resected mesothelioma discovered by self-supervised learning from 3446 whole-slide images","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5678715/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5678715/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Mesothelioma is a highly lethal and poorly biologically understood disease which presents diagnostic challenges due to its morphological complexity. This study uses self-supervised AI (Artificial Intelligence) to map the histomorphological landscape of the disease. The resulting atlas consists of recurrent patterns identified from 3446 Hematoxylin and Eosin (H\u0026E) stained images scanned from resected tumour slides. These patterns generate highly interpretable predictions, achieving state-of-the-art performance with 0.65 concordance index (c-index) for outcomes and 85% AUC in subtyping. Their clinical relevance is endorsed by comprehensive human pathological assessment. Furthermore, we characterise the molecular underpinnings of these diverse, meaningful, predictive patterns. Our approach both improves diagnosis and deepens our understanding of mesothelioma biology, highlighting the power of this self-learning method in clinical applications and scientific discovery.","manuscriptTitle":"A histomorphological atlas of resected mesothelioma discovered by self-supervised learning from 3446 whole-slide images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-17 04:02:30","doi":"10.21203/rs.3.rs-5678715/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"125f1d20-1ab8-49bd-9eb4-843f5438b9e2","owner":[],"postedDate":"January 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":42285645,"name":"Biological sciences/Cancer/Mesothelioma"},{"id":42285646,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"}],"tags":[],"updatedAt":"2025-10-08T07:05:30+00:00","versionOfRecord":{"articleIdentity":"rs-5678715","link":"https://doi.org/10.1038/s41467-025-63846-9","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2025-10-07 04:00:00","publishedOnDateReadable":"October 7th, 2025"},"versionCreatedAt":"2025-01-17 04:02:30","video":"","vorDoi":"10.1038/s41467-025-63846-9","vorDoiUrl":"https://doi.org/10.1038/s41467-025-63846-9","workflowStages":[]},"version":"v1","identity":"rs-5678715","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5678715","identity":"rs-5678715","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}