Detection of Cancer-Associated Nuclei in Histopathology Images Using Deep Convolutional Neural Networks

Detection of Cancer-Associated Nuclei in Histopathology Images Using Deep Convolutional Neural Networks | 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 Detection of Cancer-Associated Nuclei in Histopathology Images Using Deep Convolutional Neural Networks Afya Shaikh This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9260872/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 This paper presents a deep-learning pipeline for detecting nuclei-positive regions in hematoxylin and eosin (H&E) histopathology images. The study uses the public CryoNuSeg segmentation dataset, distributed through Kaggle, which contains 30 frozen tissue sections from ten human organs together with more than 8,000 manually annotated nuclei. Images were normalized, divided into smaller patches, and paired with binary masks so that the model learned a pixel-wise segmentation task rather than a slide-level diagnosis task. On the held-out evaluation set, the final model reached 0.91 pixel accuracy, with class-wise precision and recall of 0.96 and 0.92 for background pixels and 0.73 and 0.86 for nuclei pixels. The corresponding macro F1-score was 0.87, the weighted F1-score was 0.91, and the ROC–AUC score was 0.95. Qualitative comparisons showed that predicted masks tracked the overall location and shape of many nuclei-rich regions, although false positives remained in crowded or weak-contrast tissue. The results suggest that a compact convolutional neural network can provide a useful nuclei-segmentation baseline on public histopathology data and can serve as a starting point for more rigorous computational pathology work. Artificial Intelligence and Machine Learning Nuclear Medicine & Medical Imaging Deep learning Histopathology Nuclei segmentation H&E images Computational pathology Convolutional Neural Networks Full Text Additional Declarations The authors declare no competing interests. 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-9260872","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":614136750,"identity":"15b6e40c-598b-4ce9-a331-4525437f9cae","order_by":0,"name":"Afya Shaikh","email":"data:image/png;base64,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","orcid":"","institution":"Independent Researcher","correspondingAuthor":true,"prefix":"","firstName":"Afya","middleName":"","lastName":"Shaikh","suffix":""}],"badges":[],"createdAt":"2026-03-29 20:28:52","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9260872/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9260872/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105776413,"identity":"b1ebea65-3a75-40af-acde-da3458dee6b2","added_by":"auto","created_at":"2026-03-31 03:26:04","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":687242,"visible":true,"origin":"","legend":"","description":"","filename":"Shaikh2026NucleiDetectionCNN.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9260872/v1_covered_173aead1-2a86-444f-808f-ed37be2b2550.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDetection of Cancer-Associated Nuclei in Histopathology Images Using Deep Convolutional Neural Networks\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"ndependent Researcher","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":"Deep learning, Histopathology, Nuclei segmentation, H\u0026E images, Computational pathology, Convolutional Neural Networks","lastPublishedDoi":"10.21203/rs.3.rs-9260872/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9260872/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper presents a deep-learning pipeline for detecting nuclei-positive regions in hematoxylin and eosin (H\u0026amp;E) histopathology images. The study uses the public CryoNuSeg segmentation dataset, distributed through Kaggle, which contains 30 frozen tissue sections from ten human organs together with more than 8,000 manually annotated nuclei. Images were normalized, divided into smaller patches, and paired with binary masks so that the model learned a pixel-wise segmentation task rather than a slide-level diagnosis task. On the held-out evaluation set, the final model reached 0.91 pixel accuracy, with class-wise precision and recall of 0.96 and 0.92 for background pixels and 0.73 and 0.86 for nuclei pixels. The corresponding macro F1-score was 0.87, the weighted F1-score was 0.91, and the ROC–AUC score was 0.95. Qualitative comparisons showed that predicted masks tracked the overall location and shape of many nuclei-rich regions, although false positives remained in crowded or weak-contrast tissue. The results suggest that a compact convolutional neural network can provide a useful nuclei-segmentation baseline on public histopathology data and can serve as a starting point for more rigorous computational pathology work.\u003c/p\u003e","manuscriptTitle":"Detection of Cancer-Associated Nuclei in Histopathology Images Using Deep Convolutional Neural Networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-31 03:24:39","doi":"10.21203/rs.3.rs-9260872/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":"9421b10f-1026-46da-9847-d72ecb0c2fe1","owner":[],"postedDate":"March 31st, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65346049,"name":"Artificial Intelligence and Machine Learning"},{"id":65346050,"name":"Nuclear Medicine \u0026 Medical Imaging"}],"tags":[],"updatedAt":"2026-03-31T03:24:39+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-31 03:24:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9260872","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9260872","identity":"rs-9260872","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}