Multi-Task Learning Model for Enhancing Tumor-Infiltrating Lymphocyte Prediction

Multi-Task Learning Model for Enhancing Tumor-Infiltrating Lymphocyte Prediction | 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 Multi-Task Learning Model for Enhancing Tumor-Infiltrating Lymphocyte Prediction Sijin Kim, Kazi Rakib Hasan, Seokhwan Ko, Ji Young Park, Hyung Soo Han, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7952473/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Quantifying tumor-infiltrating lymphocytes (TILs) from histopathology images provides essential insights into tumor–immune interactions and serves as a key biomarker for prognosis and immunotherapy response in breast cancer. Conventional computational approaches require separate tissue segmentation and lymphocyte detection, which leads to redundant computation and limited information sharing between related features. To address these challenges, we propose a unified multi-task learning (MTL) framework that simultaneously performs tissue segmentation and lymphocyte detection to enhance TIL prediction accuracy and efficiency. The proposed model is built upon a pathology-specific large vision model (LVM) pretrained through self-supervised learning and incorporates layer-wise embedding sharing to enhance deep cross-task feature interactions. Experimental results on the TIGER Challenge dataset demonstrated that the proposed unified MTL model achieved improved TIL prediction accuracy compared to both single-task and conventional MTL models. In particular, compared to the conventional MTL model, the proposed approach achieved a 7.0% improvement in Pearson correlation and a 5.9% increase in Spearman correlation, while reducing the number of trainable parameters by approximately 33%, demonstrating superior performance in both predictive accuracy and computational efficiency. When benchmarked against top-performing models from the TIGER Challenge, the proposed framework achieved comparable or higher accuracy across tissue segmentation, lymphocyte detection, and TIL prediction metrics. The framework also improved concordance between predicted and ground-truth TIL scores across multiple correlation metrics while maintaining robust segmentation and detection performance. These results highlight that joint optimization of tissue- and cell-level tasks can yield biologically coherent representations of the tumor microenvironment, supporting automated TIL quantification. The proposed MTL approach provides a scalable framework with potential applicability in digital pathology and computational oncology. Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Tumor-infiltrating lymphocyte Tissue segmentation and lymphocytes detection Multi-task learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted 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-7952473","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":551362606,"identity":"0ae72443-56f0-480a-9e94-4aa5b83c5550","order_by":0,"name":"Sijin Kim","email":"","orcid":"","institution":"Kyungpook National University","correspondingAuthor":false,"prefix":"","firstName":"Sijin","middleName":"","lastName":"Kim","suffix":""},{"id":551362607,"identity":"c2641588-e8c1-4087-a791-859ab1351f58","order_by":1,"name":"Kazi Rakib Hasan","email":"","orcid":"","institution":"Kyungpook National University","correspondingAuthor":false,"prefix":"","firstName":"Kazi","middleName":"Rakib","lastName":"Hasan","suffix":""},{"id":551362608,"identity":"e7207e47-3b58-4803-87ca-d280f3952d7b","order_by":2,"name":"Seokhwan Ko","email":"","orcid":"","institution":"Kyungpook National University","correspondingAuthor":false,"prefix":"","firstName":"Seokhwan","middleName":"","lastName":"Ko","suffix":""},{"id":551362609,"identity":"57582b14-942a-4f68-9e85-068c4ee73044","order_by":3,"name":"Ji Young Park","email":"","orcid":"","institution":"Kyungpook National University School of Medicine, Kyungpook National University Chilgok Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ji","middleName":"Young","lastName":"Park","suffix":""},{"id":551362610,"identity":"1b13da21-5d1a-4d4d-8e4a-c36e546e456b","order_by":4,"name":"Hyung Soo Han","email":"","orcid":"","institution":"Kyungpook National University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hyung","middleName":"Soo","lastName":"Han","suffix":""},{"id":551362611,"identity":"e35be2e7-a3b5-47a5-b6e6-726caa44ef54","order_by":5,"name":"Junghwan Cho","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYBACxgYGNiBlYwDlJxCtJY0ELUAA0nKYBC3M7e3PHvPUnDc2l0hg/PCDIS2fsMN6zpgb8xy7bWY5I4FZsochx7KBoJYZOWzSPGy3bQxuJDBIMzBUGBDSAdSS/kya5985kBbm30RqSTCT5m07YAbUwga0JYcILT1nzCTn9iUbG5x52GbZY5BGWIshMMQk3nyzM9xwPPnwjR8VyURoaUBYCGQS1sDAIE+EmlEwCkbBKBjpAAAUfzYSG5VGXAAAAABJRU5ErkJggg==","orcid":"","institution":"Kyungpook National University","correspondingAuthor":true,"prefix":"","firstName":"Junghwan","middleName":"","lastName":"Cho","suffix":""}],"badges":[],"createdAt":"2025-10-27 11:27:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7952473/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7952473/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97192320,"identity":"b29f22fb-7891-4c17-8126-abcc9b4a9c0b","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8206,"visible":true,"origin":"","legend":"","description":"","filename":"ee6bed1f382a4f0a8dca1399d63ab33b.json","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/64c8a80ba09d8ae444aed1d6.json"},{"id":97250763,"identity":"732dcdbb-1eb6-466e-bdbe-a67764789bbd","added_by":"auto","created_at":"2025-12-02 13:15:10","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1149264,"visible":true,"origin":"","legend":"","description":"","filename":"MTLSR102725r.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/e2d776e948fc41bd15445ce4.pdf"},{"id":97192323,"identity":"0f31ec50-a161-4a6e-ade3-40631012e51a","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":186570,"visible":true,"origin":"","legend":"","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/8ed0523ed1bc974d4393f148.png"},{"id":97192321,"identity":"615d9542-09ac-4775-be36-7bfd023eb037","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127794,"visible":true,"origin":"","legend":"","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/578699def14616283aa88443.png"},{"id":97192322,"identity":"97404c42-7601-42de-a7ea-f711c924d08c","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":726920,"visible":true,"origin":"","legend":"","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/e9c9be210f3277701d18b36b.png"},{"id":97249336,"identity":"8dfa9330-68c4-4ef0-8cb2-0ec7c2bd1297","added_by":"auto","created_at":"2025-12-02 13:12:13","extension":"ldf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49107,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaall.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/9cd0a545a9628960e52f9f48.ldf"},{"id":97249515,"identity":"0d0027db-abd6-414f-afed-e3deb32fa59f","added_by":"auto","created_at":"2025-12-02 13:12:47","extension":"ldf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268657,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrvltwaen.ldf","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/97e7df913377932ed7a9e45c.ldf"},{"id":97192325,"identity":"73c30a8f-dd4a-4775-a0d5-7caa0a63a821","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"sty","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15480,"visible":true,"origin":"","legend":"","description":"","filename":"jabbrv.sty","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/a8d210e3a0dade6609df6aec.sty"},{"id":97249358,"identity":"1c70ddd7-ee8e-4279-933e-21ffa0e48916","added_by":"auto","created_at":"2025-12-02 13:12:20","extension":"bst","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36153,"visible":true,"origin":"","legend":"","description":"","filename":"naturemagdoi.bst","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/f17d2f1f18f83885b786e54e.bst"},{"id":97192328,"identity":"be9e661f-c0ed-4a1c-ab95-fdebaeac8eb4","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":379752,"visible":true,"origin":"","legend":"","description":"","filename":"stream.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/6a9ff2f40e79ce066ec8bafe.jpg"},{"id":97192330,"identity":"84d5284b-4fe0-49ca-ad63-a98da6f6f1d5","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"cls","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5815,"visible":true,"origin":"","legend":"","description":"","filename":"wlscirep.cls","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/21e231bfde9cc91fa4a43ec4.cls"},{"id":97192326,"identity":"50b44d69-d636-4f93-8d3b-a05461d1762e","added_by":"auto","created_at":"2025-12-01 20:15:21","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":57646,"visible":true,"origin":"","legend":"","description":"","filename":"ee6bed1f382a4f0a8dca1399d63ab33b1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1/43a56f86e14f70dcd3a5465f.xml"},{"id":105704440,"identity":"ff49911d-baac-4ec9-9bb8-143407b99e28","added_by":"auto","created_at":"2026-03-30 06:43:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":453053,"visible":true,"origin":"","legend":"","description":"","filename":"MTLSR102725r.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7952473/v1_covered_9c601b84-f17d-4070-b0c4-58af047744b7.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multi-Task Learning Model for Enhancing Tumor-Infiltrating Lymphocyte Prediction","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":true,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Tumor-infiltrating lymphocyte, Tissue segmentation and lymphocytes detection, Multi-task learning","lastPublishedDoi":"10.21203/rs.3.rs-7952473/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7952473/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Quantifying tumor-infiltrating lymphocytes (TILs) from histopathology images provides essential insights into tumor–immune interactions and serves as a key biomarker for prognosis and immunotherapy response in breast cancer. Conventional computational approaches require separate tissue segmentation and lymphocyte detection, which leads to redundant computation and limited information sharing between related features. To address these challenges, we propose a unified multi-task learning (MTL) framework that simultaneously performs tissue segmentation and lymphocyte detection to enhance TIL prediction accuracy and efficiency. The proposed model is built upon a pathology-specific large vision model (LVM) pretrained through self-supervised learning and incorporates layer-wise embedding sharing to enhance deep cross-task feature interactions. Experimental results on the TIGER Challenge dataset demonstrated that the proposed unified MTL model achieved improved TIL prediction accuracy compared to both single-task and conventional MTL models. In particular, compared to the conventional MTL model, the proposed approach achieved a 7.0% improvement in Pearson correlation and a 5.9% increase in Spearman correlation, while reducing the number of trainable parameters by approximately 33%, demonstrating superior performance in both predictive accuracy and computational efficiency. When benchmarked against top-performing models from the TIGER Challenge, the proposed framework achieved comparable or higher accuracy across tissue segmentation, lymphocyte detection, and TIL prediction metrics. The framework also improved concordance between predicted and ground-truth TIL scores across multiple correlation metrics while maintaining robust segmentation and detection performance. These results highlight that joint optimization of tissue- and cell-level tasks can yield biologically coherent representations of the tumor microenvironment, supporting automated TIL quantification. The proposed MTL approach provides a scalable framework with potential applicability in digital pathology and computational oncology.","manuscriptTitle":"Multi-Task Learning Model for Enhancing Tumor-Infiltrating Lymphocyte Prediction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-01 20:15:16","doi":"10.21203/rs.3.rs-7952473/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a1af9ad0-86cf-47a1-ac78-2da55e3499e4","owner":[],"postedDate":"December 1st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":58672340,"name":"Biological sciences/Cancer"},{"id":58672341,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":58672342,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-03-30T06:42:48+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-01 20:15:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7952473","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7952473","identity":"rs-7952473","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}