Interpretable and Robust Deep Learning for Automated HER2 Assessment in Breast Cancer

Interpretable and Robust Deep Learning for Automated HER2 Assessment in Breast Cancer | 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 Interpretable and Robust Deep Learning for Automated HER2 Assessment in Breast Cancer Md Serajun Nabi, Thinley Yeshey Choden, S M Asiful Islam Saky, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8867647/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 14 You are reading this latest preprint version Abstract Accurate determination of human epidermal growth factor receptor 2 (HER2) status is essential for guiding targeted therapy in breast cancer. Yet, manual immunohistochemistry (IHC) scoring remains susceptible to inter-observer variability, particularly in borderline cases. Although deep learning–based methods have shown promise for automated HER2 assessment, their clinical adoption is hindered by limited interpretability, poor robustness across imaging magnifications, and insufficient alignment with pathological reasoning. In this study, we propose an interpretable and computationally efficient deep learning framework for automated HER2 scoring that operates consistently across tissue-level (10×) and cellular-level (40×) histopathological images. The framework employs a hybrid layer unfreezing strategy to balance feature adaptation and computational cost, enabling robust multi-magnification learning without reliance on extensive fine-tuning. To address the limitations of qualitative explainability, we integrate Score-CAM with quantitatively validated, membrane-focused metrics Membrane Activation Precision (MAP) and Explanation Consistency (EC) to objectively assess the clinical relevance and stability of model explanations against pathologist annotations. The proposed approach is evaluated on three public HER2 IHC datasets, including BCI, HER2-IHC-40x-Patch, and HER2-IHC-40x-WSI. Experimental results demonstrate strong and consistent performance across magnifications, achieving up to 96% accuracy on high-resolution patches and maintaining robust performance at 10× magnification, with near-perfect discrimination of HER2 3+ cases (AUC > 0.99). Furthermore, the framework reduces computational overhead compared to full fine-tuning while improving cross-magnification generalizability relative to existing methods. By combining robust multi-scale performance with clinically grounded explainability, this work advances trustworthy AI-assisted HER2 scoring and addresses key barriers to the deployment of automated decision-support systems in digital pathology. Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Health sciences/Oncology Full Text Additional Declarations No competing interests reported. Supplementary Files ScientificReporther2Paper2.zip Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Mar, 2026 Reviews received at journal 13 Mar, 2026 Reviews received at journal 10 Mar, 2026 Reviewers agreed at journal 09 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviews received at journal 05 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviewers invited by journal 05 Mar, 2026 Editor assigned by journal 25 Feb, 2026 Editor invited by journal 20 Feb, 2026 Submission checks completed at journal 18 Feb, 2026 First submitted to journal 18 Feb, 2026 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-8867647","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":603289500,"identity":"533e4c68-c63b-4aac-87f4-94a0f718df7f","order_by":0,"name":"Md Serajun Nabi","email":"","orcid":"","institution":"Multimedia University","correspondingAuthor":false,"prefix":"","firstName":"Md","middleName":"Serajun","lastName":"Nabi","suffix":""},{"id":603289501,"identity":"02cd5738-9277-4270-aea0-850f5b77157b","order_by":1,"name":"Thinley Yeshey Choden","email":"","orcid":"","institution":"Albukhary International University","correspondingAuthor":false,"prefix":"","firstName":"Thinley","middleName":"Yeshey","lastName":"Choden","suffix":""},{"id":603289502,"identity":"b26950d3-4506-4f37-bf09-99da9424dd55","order_by":2,"name":"S M Asiful Islam Saky","email":"","orcid":"","institution":"Albukhary International University","correspondingAuthor":false,"prefix":"","firstName":"S","middleName":"M Asiful Islam","lastName":"Saky","suffix":""},{"id":603289503,"identity":"49f3db49-778d-4af4-9877-6b79ec31055e","order_by":3,"name":"Dema Yuden","email":"","orcid":"","institution":"Albukhary International University","correspondingAuthor":false,"prefix":"","firstName":"Dema","middleName":"","lastName":"Yuden","suffix":""},{"id":603289504,"identity":"61171f47-67d9-4a49-8d9d-f1dc04143302","order_by":4,"name":"Mohammad Faizal Ahmad Fauzi","email":"","orcid":"","institution":"KPJ healthcare University","correspondingAuthor":false,"prefix":"","firstName":"Mohammad","middleName":"Faizal Ahmad","lastName":"Fauzi","suffix":""},{"id":603289505,"identity":"36861d02-2847-497f-9ea0-4aef8b928970","order_by":5,"name":"Hezerul Bin Karim","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuUlEQVRIiWNgGAWjYBACAzBZYAPh8RCvxSCNdC2HSdBizt7+8HGFwflo3RkJjA/etjFEGxwgoMWy50Cy4RmD27nbbiQwG85tY8jdQEiLwY2EY5INEC1s0rxEabn/sP1ng8E5kBb238RpucHMxthgcABsCzNxWs6kMQMdlpy77czDZsk55yRyZxLUcvz4w48NFXa5244nH/zwpswmt4+QFiTA2AAkJBgUSNACBfINJGsZBaNgFIyCYQ4AAlJH0cx6NXAAAAAASUVORK5CYII=","orcid":"","institution":"Multimedia University","correspondingAuthor":true,"prefix":"","firstName":"Hezerul","middleName":"Bin","lastName":"Karim","suffix":""}],"badges":[],"createdAt":"2026-02-13 05:23:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8867647/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8867647/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104405882,"identity":"3d37a3b9-07ce-4893-a60c-349b5efcc783","added_by":"auto","created_at":"2026-03-11 12:24:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2462264,"visible":true,"origin":"","legend":"","description":"","filename":"her2scientirepo.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8867647/v1_covered_53896237-c533-45bb-9a58-ebe8e66169f9.pdf"},{"id":104348932,"identity":"5c28e5c3-0b93-4f87-a9cc-b74aed1f16ea","added_by":"auto","created_at":"2026-03-10 18:40:03","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":9588649,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReporther2Paper2.zip","url":"https://assets-eu.researchsquare.com/files/rs-8867647/v1/63f103dbc27aff64dcf84696.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"Interpretable and Robust Deep Learning for Automated HER2 Assessment in Breast Cancer","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"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-8867647/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8867647/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Accurate determination of human epidermal growth factor receptor 2 (HER2) status is essential for guiding targeted therapy in breast cancer. Yet, manual immunohistochemistry (IHC) scoring remains susceptible to inter-observer variability, particularly in borderline cases. Although deep learning–based methods have shown promise for automated HER2 assessment, their clinical adoption is hindered by limited interpretability, poor robustness across imaging magnifications, and insufficient alignment with pathological reasoning. In this study, we propose an interpretable and computationally efficient deep learning framework for automated HER2 scoring that operates consistently across tissue-level (10×) and cellular-level (40×) histopathological images. The framework employs a hybrid layer unfreezing strategy to balance feature adaptation and computational cost, enabling robust multi-magnification learning without reliance on extensive fine-tuning. To address the limitations of qualitative explainability, we integrate Score-CAM with quantitatively validated, membrane-focused metrics Membrane Activation Precision (MAP) and Explanation Consistency (EC) to objectively assess the clinical relevance and stability of model explanations against pathologist annotations. The proposed approach is evaluated on three public HER2 IHC datasets, including BCI, HER2-IHC-40x-Patch, and HER2-IHC-40x-WSI. Experimental results demonstrate strong and consistent performance across magnifications, achieving up to 96% accuracy on high-resolution patches and maintaining robust performance at 10× magnification, with near-perfect discrimination of HER2 3+ cases (AUC \u003e 0.99). Furthermore, the framework reduces computational overhead compared to full fine-tuning while improving cross-magnification generalizability relative to existing methods. By combining robust multi-scale performance with clinically grounded explainability, this work advances trustworthy AI-assisted HER2 scoring and addresses key barriers to the deployment of automated decision-support systems in digital pathology.","manuscriptTitle":"Interpretable and Robust Deep Learning for Automated HER2 Assessment in Breast Cancer","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-10 18:39:57","doi":"10.21203/rs.3.rs-8867647/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-17T11:11:52+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-13T05:18:06+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-10T17:35:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"5871326178848960034150784001575995590","date":"2026-03-09T16:24:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6354464008681476862263492376308928967","date":"2026-03-06T00:14:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-05T17:35:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"41649389006754229205654141792669755468","date":"2026-03-05T16:54:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"145249017568288580825051444768007201847","date":"2026-03-05T12:46:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"334250566750614524124777714761217002258","date":"2026-03-05T12:16:23+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-05T11:52:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-25T07:35:22+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-20T06:50:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-18T09:04:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-18T09:01:01+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":"c51b99d3-675e-41da-838d-714a9a67052f","owner":[],"postedDate":"March 10th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64195985,"name":"Biological sciences/Cancer"},{"id":64195986,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":64195987,"name":"Physical sciences/Mathematics and computing"},{"id":64195988,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2026-05-10T07:10:11+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-10 18:39:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8867647","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8867647","identity":"rs-8867647","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}