Impact of Pre-processing and Local Feature Extraction on Feature-Based Registration of Whole-Slide Images

Impact of Pre-processing and Local Feature Extraction on Feature-Based Registration of 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 Research Article Impact of Pre-processing and Local Feature Extraction on Feature-Based Registration of Whole-Slide Images Arthur Elskens, Adrien Foucart, Olivier Debeir, Christine Decaestecker This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6678947/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 Feature-based registration has gained increasing popularity in digital pathology as a means of achieving initial global, low-resolution alignment between image pairs. Despite its widespread adoption, the specific design choices within registration pipelines are often insufficiently justified. This study presents a comprehensive benchmarking analysis on consecutive multi-stained whole-slide images to evaluate the performance of various pre-processing and local feature extraction methods. Both traditional and deep learning-based techniques are assessed to determine whether the latter consistently outperform the former. The findings underscore the critical importance of both pre-processing and feature description steps in influencing overall alignment quality. Notably, Grayscale conversion consistently surpasses Hematoxylin deconvolution as a pre-processing approach. While detector selection has a relatively minor impact on performance, descriptor choice plays a crucial role. Among the most robust descriptors identified are two deep learning-based methods ( SuperPoint and DISK ), as well as two classical algorithms ( BRIEF and O-BRIEF ), which deliver competitive results with lower computational demands. In more challenging registration scenarios, however, SuperPoint emerges as the most effective descriptor. Biomedical Engineering Digital pathology Whole-slide image Feature-based registration Benchmarking 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-6678947","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":457504503,"identity":"050ef4f0-05f0-487d-835b-39e7b5bb5557","order_by":0,"name":"Arthur Elskens","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIiWNgGAWjYHACNjDJB6FtGPgkoCK4ATMbVCOYTmNgI1XLYcJa+GefP/aYdw+DHBv/scTHBX/OJ7ZJN7A9+IBHi8S5ZHZjnmcMxmwSaYeNZ7bdTmyTOcBuOAOfNWeY2aR5DjAktkmwt0nzNgC1SCQARfDokIdqqW/jP97+m+fPOYiWP3i0GEC1JLAxpB1j5mE7ANGCz12GZ5jNJOcckDBsk0hLluZtSzZukznYbtiDR4vcGcZnEm8O2Mjz8x8z/Mzzx062X7r52IMf+KyBAAlkDmMDYQ2jYBSMglEwCvACAHZxQEFEQCF3AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0001-2288-464X","institution":"Université Libre de Bruxelles","correspondingAuthor":true,"prefix":"","firstName":"Arthur","middleName":"","lastName":"Elskens","suffix":""},{"id":457504786,"identity":"972356bd-9c2c-41af-ba79-3c838a2d2e37","order_by":1,"name":"Adrien Foucart","email":"","orcid":"https://orcid.org/0000-0002-7280-0352","institution":"Université Libre de Bruxelles","correspondingAuthor":false,"prefix":"","firstName":"Adrien","middleName":"","lastName":"Foucart","suffix":""},{"id":457505150,"identity":"9431d5e3-1f1e-42c4-bcfa-c779472cb7a3","order_by":2,"name":"Olivier Debeir","email":"","orcid":"","institution":"Université Libre de Bruxelles","correspondingAuthor":false,"prefix":"","firstName":"Olivier","middleName":"","lastName":"Debeir","suffix":""},{"id":457505151,"identity":"cf5c0918-dbad-4263-9c70-72c4d03533a8","order_by":3,"name":"Christine Decaestecker","email":"","orcid":"","institution":"Université Libre de Bruxelles","correspondingAuthor":false,"prefix":"","firstName":"Christine","middleName":"","lastName":"Decaestecker","suffix":""}],"badges":[],"createdAt":"2025-05-16 08:39:13","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-6678947/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6678947/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83030981,"identity":"6bf4d011-d22b-48bf-bc24-a6aa60582cd0","added_by":"auto","created_at":"2025-05-19 09:08:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6390756,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6678947/v1_covered_0d8a1885-b2fe-4ad5-befd-5fa79add076b.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eImpact of Pre-processing and Local Feature Extraction on Feature-Based Registration of Whole-Slide Images\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Université Libre de Bruxelles","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":"Digital pathology, Whole-slide image, Feature-based registration, Benchmarking","lastPublishedDoi":"10.21203/rs.3.rs-6678947/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6678947/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eFeature-based registration has gained increasing popularity in digital pathology as a means of achieving initial global, low-resolution alignment between image pairs. Despite its widespread adoption, the specific design choices within registration pipelines are often insufficiently justified. This study presents a comprehensive benchmarking analysis on consecutive multi-stained whole-slide images to evaluate the performance of various pre-processing and local feature extraction methods. Both traditional and deep learning-based techniques are assessed to determine whether the latter consistently outperform the former. The findings underscore the critical importance of both pre-processing and feature description steps in influencing overall alignment quality. Notably, \u003cem\u003eGrayscale conversion\u003c/em\u003e consistently surpasses \u003cem\u003eHematoxylin deconvolution\u003c/em\u003e as a pre-processing approach. While detector selection has a relatively minor impact on performance, descriptor choice plays a crucial role. Among the most robust descriptors identified are two deep learning-based methods (\u003cem\u003eSuperPoint\u003c/em\u003e and \u003cem\u003eDISK\u003c/em\u003e), as well as two classical algorithms (\u003cem\u003eBRIEF\u003c/em\u003e and \u003cem\u003eO-BRIEF\u003c/em\u003e), which deliver competitive results with lower computational demands. In more challenging registration scenarios, however, \u003cem\u003eSuperPoint\u003c/em\u003e emerges as the most effective descriptor.\u003c/p\u003e","manuscriptTitle":"Impact of Pre-processing and Local Feature Extraction on Feature-Based Registration of Whole-Slide Images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-19 09:00:03","doi":"10.21203/rs.3.rs-6678947/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":"7bb371ba-4e0c-423f-9b90-46f37c91ce09","owner":[],"postedDate":"May 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48623227,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2025-05-19T09:00:03+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-19 09:00:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6678947","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6678947","identity":"rs-6678947","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}