Graph Neural Networks in Multi-Stained Pathological Imaging: Extended Comparative Analysis of Radiomic Features

Graph Neural Networks in Multi-Stained Pathological Imaging: Extended Comparative Analysis of Radiomic Features | 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 Graph Neural Networks in Multi-Stained Pathological Imaging: Extended Comparative Analysis of Radiomic Features Luis Carlos Rivera Monroy, Leonhard Rist, Christian Ostalecki, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4241891/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 Purpose: This study investigates Radiomics features for Graph Neural Networks (GNNs) in MELC pathology sample classification focusing on often misdiagnosed skin diseases. Methods: GNNs processing multiple pathological slides together as cell-level graphs are compared to XGBoost and Random Forest. The analysis assesses the use of MELC vs. Radiomics features, their dimensionality reduction with UMAP or tSNE and the graph connectivity based on spatial and feature closeness. Results: Integrating Radiomics features in a spatially connected graph markedly outperforms standard models when classifying pathologically similar diseases. Additionally, the UMAP dimensionality reduction techniques improves GNN classification performance. Conclusion: Radiomics, processed with GNNs, shows promise for multi-disease classification, enhancing diagnosis accuracy. Considering the potential, future research should extend these methods to a broader range of diseases. Pathology Dermatology Artificial Intelligence and Machine Learning Radiomics Multi-array Imaging Graph Neural Network MELC Histopathological Analysis 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-4241891","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":289284655,"identity":"4fad7de8-e394-40f1-8c7e-3baad9c0663c","order_by":0,"name":"Luis Carlos Rivera Monroy","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0UlEQVRIiWNgGAWjYBACPgbGBiAlAWIzPqgAUQcIaGFD0sJscIY4LUhsCeK0SCQ3f2DcY5HYL334WcWBmsMMfMcbCGlJbJNgeCaROLMvzezGgWOHGSTPELAGpAXoFgljoD/Mbn9gO8xgcCOBoJbmDyAt9mfYvxUc+AfUcv8BQS0NEkAtcgY8PGYMB9tAtuDXwcDG87BNIgGoReIMT7HEwb50HskzBBzGz57++MOHA3U8/D3sGz8c+GYtx3f8AAFrQADZWB4i1I+CUTAKRsEoIAQA3ZNCQ+br9RkAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-8232-8920","institution":"Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":true,"prefix":"","firstName":"Luis","middleName":"Carlos Rivera","lastName":"Monroy","suffix":""},{"id":289284656,"identity":"0075848b-0961-440b-94bf-c6d80d0ea90b","order_by":1,"name":"Leonhard Rist","email":"","orcid":"","institution":"Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Leonhard","middleName":"","lastName":"Rist","suffix":""},{"id":289284657,"identity":"d195e1b5-5fd8-4404-930a-b350a2d991de","order_by":2,"name":"Christian Ostalecki","email":"","orcid":"","institution":"Department of Dermatology, Universitätsklinikum Erlangen","correspondingAuthor":false,"prefix":"","firstName":"Christian","middleName":"","lastName":"Ostalecki","suffix":""},{"id":289284658,"identity":"f3984f63-559a-4b68-bb59-b8cea3390bce","order_by":3,"name":"Andreas Bauer","email":"","orcid":"","institution":"Department of Dermatology, Universitätsklinikum Erlangen","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Bauer","suffix":""},{"id":289284659,"identity":"7541dbc1-3139-466d-94f8-b04cb7a5697c","order_by":4,"name":"Julio Vera","email":"","orcid":"","institution":"Department of Dermatology, Universitätsklinikum Erlangen","correspondingAuthor":false,"prefix":"","firstName":"Julio","middleName":"","lastName":"Vera","suffix":""},{"id":289284660,"identity":"84044f2a-8bcf-4f2c-92a8-5107c99c4556","order_by":5,"name":"Katharina Breininger","email":"","orcid":"","institution":"Department of Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Katharina","middleName":"","lastName":"Breininger","suffix":""},{"id":289284661,"identity":"3e560933-e003-40e9-a202-5078310ac794","order_by":6,"name":"Andreas Maier","email":"","orcid":"","institution":"Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Maier","suffix":""}],"badges":[],"createdAt":"2024-04-09 12:31:21","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4241891/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4241891/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":54403920,"identity":"25248ddc-70d2-42cb-8b68-0fe4c1fee365","added_by":"auto","created_at":"2024-04-10 03:05:43","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":613526,"visible":true,"origin":"","legend":"","description":"","filename":"snarticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4241891/v1_covered_886dd9b5-6f65-457f-bec1-560137ab0995.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eGraph Neural Networks in Multi-Stained Pathological Imaging: Extended Comparative Analysis of Radiomic Features\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Erlangen-Nuremberg","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":"Radiomics, Multi-array Imaging, Graph Neural Network, MELC, Histopathological Analysis","lastPublishedDoi":"10.21203/rs.3.rs-4241891/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4241891/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose:\u003c/strong\u003e This study investigates Radiomics features for Graph Neural Networks (GNNs) in MELC pathology sample classification focusing on often misdiagnosed skin diseases.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eGNNs processing multiple pathological slides together as cell-level graphs are compared to XGBoost and Random Forest. The analysis assesses the use of MELC vs. Radiomics features, their dimensionality reduction with UMAP or tSNE and the graph connectivity based on spatial and feature closeness.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Integrating Radiomics features in a spatially connected graph markedly outperforms standard models when classifying pathologically similar diseases. Additionally, the UMAP dimensionality reduction techniques improves GNN classification performance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e Radiomics, processed with GNNs, shows promise for multi-disease classification, enhancing diagnosis accuracy. Considering the potential, future research should extend these methods to a broader range of diseases.\u003c/p\u003e","manuscriptTitle":"Graph Neural Networks in Multi-Stained Pathological Imaging: Extended Comparative Analysis of Radiomic Features","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-10 02:57:35","doi":"10.21203/rs.3.rs-4241891/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":"d0f1afd7-772c-4ade-8137-3c5b01f0d72a","owner":[],"postedDate":"April 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":30468334,"name":"Pathology"},{"id":30468335,"name":"Dermatology"},{"id":30468336,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2024-04-10T02:57:35+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-10 02:57:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4241891","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4241891","identity":"rs-4241891","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}