Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography

Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography | 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 Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography Ekin Yagis, Shahab Aslani, Yashvardhan Jain, Yang Zhou, Shahrokh Rahmani, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4613439/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Nov, 2024 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Automated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT). We begin with an extensive review of current machine learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using Hierarchical Phase-Contrast Tomography (HiP-CT) as part of the Human Organ Atlas Project. HiP-CT, pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering resolution around 20µm/voxel, and enabling highly detailed localized zooms up to 1µm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging. Our three experiments yielded Dice scores of 0.9523 and 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores—such as centerline Dice values ranging from 0.82 to 0.88, certain errors persisted. Specifically, large vessels that collapsed due to the lack of hydro-static pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Through our review and outputs, we aim to set a benchmark for subsequent model evaluations using various modalities, especially with the HiP-CT imaging database. Physical sciences/Mathematics and computing/Computer science Health sciences/Anatomy/Kidney Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 08 Nov, 2024 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 16 Aug, 2024 Reviews received at journal 13 Aug, 2024 Reviewers agreed at journal 03 Aug, 2024 Reviews received at journal 02 Aug, 2024 Reviewers agreed at journal 28 Jul, 2024 Reviewers invited by journal 23 Jul, 2024 Editor assigned by journal 23 Jul, 2024 Editor invited by journal 25 Jun, 2024 Submission checks completed at journal 24 Jun, 2024 First submitted to journal 20 Jun, 2024 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-4613439","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":326316055,"identity":"8a36b3b5-367f-4cb8-83e8-0f77311de936","order_by":0,"name":"Ekin Yagis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYNACAwY5BgbmBjBbglgtxgwMjCRpYWBIbCBaCz8Dj+HHLwV16RuOH2xg+FHDkDizgYAWyQYeY2kZg8O5G84kNjD2HGNInE3IFoMDPAbSEgYHcjfcADqMt4EhcR4hLfYHeIx/SxjUpRsAtTD+JUaLAQOPmeQHA+YEkBZmkC0EHSZxmK3MmsHgsOFMoF8OyxyTMCboff725s03f/ypk+c7fvjgwzc1NrIzDhCyhpnDgJkHyj5AZESyP2D8QYy6UTAKRsEoGLkAAILNPReh7Pa+AAAAAElFTkSuQmCC","orcid":"","institution":"University College London","correspondingAuthor":true,"prefix":"","firstName":"Ekin","middleName":"","lastName":"Yagis","suffix":""},{"id":326316056,"identity":"d137f229-90fa-42fa-a458-c0ae03c67680","order_by":1,"name":"Shahab Aslani","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Shahab","middleName":"","lastName":"Aslani","suffix":""},{"id":326316057,"identity":"aa235346-5209-4063-854e-2f2cec914081","order_by":2,"name":"Yashvardhan Jain","email":"","orcid":"","institution":"Indiana University Bloomington","correspondingAuthor":false,"prefix":"","firstName":"Yashvardhan","middleName":"","lastName":"Jain","suffix":""},{"id":326316058,"identity":"9e6d1a21-2363-4217-9550-1dba052b2331","order_by":3,"name":"Yang Zhou","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Zhou","suffix":""},{"id":326316059,"identity":"e54ba1d4-c572-4e15-8c83-04947919b8d9","order_by":4,"name":"Shahrokh Rahmani","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Shahrokh","middleName":"","lastName":"Rahmani","suffix":""},{"id":326316060,"identity":"1e94d9bd-15ca-48cf-bccc-07439582f9e5","order_by":5,"name":"Joseph Brunet","email":"","orcid":"","institution":"European Synchrotron Radiation Facility","correspondingAuthor":false,"prefix":"","firstName":"Joseph","middleName":"","lastName":"Brunet","suffix":""},{"id":326316061,"identity":"05ca48ea-328c-47fb-8716-85cbacc14069","order_by":6,"name":"Alexandre Bellier","email":"","orcid":"","institution":"Laboratoire d’Anatomie Des Alpes Francaises","correspondingAuthor":false,"prefix":"","firstName":"Alexandre","middleName":"","lastName":"Bellier","suffix":""},{"id":326316062,"identity":"6dea2b25-7721-40eb-a8b2-298a28c63456","order_by":7,"name":"Christopher Werlein","email":"","orcid":"","institution":"Institute of Pathology, Hannover Medical School, Carl-Neuberg-Straße 1, 30625, Hannover, Germany","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Werlein","suffix":""},{"id":326316063,"identity":"65f4f76c-f787-4567-9523-7590f9e19626","order_by":8,"name":"Maximilian Ackermann","email":"","orcid":"","institution":"University Medical Center of the Johannes Gutenberg University Mainz","correspondingAuthor":false,"prefix":"","firstName":"Maximilian","middleName":"","lastName":"Ackermann","suffix":""},{"id":326316064,"identity":"0b8a3535-b22d-497d-a7ff-08cfecdd9bee","order_by":9,"name":"Danny Jonigk","email":"","orcid":"","institution":"Member of the German Center for Lung Research (DZL), Biomedical Research in Endstage and Obstructive Lung Disease Hannover (BREATH), Hannover","correspondingAuthor":false,"prefix":"","firstName":"Danny","middleName":"","lastName":"Jonigk","suffix":""},{"id":326316065,"identity":"307a98b9-b239-4ae4-9aab-15cfd407e0a6","order_by":10,"name":"Paul Tafforeau","email":"","orcid":"","institution":"European Synchrotron Radiation Facility","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"","lastName":"Tafforeau","suffix":""},{"id":326316066,"identity":"0b77bbaa-21e5-4bcd-ab77-3cc9e64709a0","order_by":11,"name":"Claire Walsh","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"","lastName":"Walsh","suffix":""},{"id":326316067,"identity":"b0ec993b-cf0d-4921-b051-173d1e157e3c","order_by":12,"name":"Peter D Lee","email":"","orcid":"","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"D","lastName":"Lee","suffix":""}],"badges":[],"createdAt":"2024-06-20 18:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4613439/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4613439/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-024-77582-5","type":"published","date":"2024-11-08T15:57:17+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":68749882,"identity":"022f8f48-d141-4198-9082-a6fd60086b2a","added_by":"auto","created_at":"2024-11-11 16:07:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1674985,"visible":true,"origin":"","legend":"","description":"","filename":"FinalsubmissionstoScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4613439/v1_covered_c124a4e5-ad4f-4f24-94ad-8302c5d5fcf9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"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-4613439/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4613439/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAutomated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT).\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe begin with an extensive review of current machine learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using Hierarchical Phase-Contrast Tomography (HiP-CT) as part of the Human Organ Atlas Project. HiP-CT, pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering resolution around 20µm/voxel, and enabling highly detailed localized zooms up to 1µm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging.\u003c/p\u003e\n\u003cp\u003eOur three experiments yielded Dice scores of 0.9523 and 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores—such as centerline Dice values ranging from 0.82 to 0.88, certain errors persisted.\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSpecifically, large vessels that collapsed due to the lack of hydro-static pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Through our review and outputs, we aim to set a benchmark for subsequent model evaluations using various modalities, especially with the HiP-CT imaging database.\u003c/p\u003e","manuscriptTitle":"Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-16 02:17:02","doi":"10.21203/rs.3.rs-4613439/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-08-16T06:13:10+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-08-13T20:33:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"278023517879373982742221115460670050755","date":"2024-08-03T19:32:45+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-08-02T08:01:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"211128328964057025154266871042085262850","date":"2024-07-28T07:32:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-24T00:15:05+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-23T14:39:53+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-06-25T12:32:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-24T15:29:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-06-20T18:22:11+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":"f1a517d8-6207-4196-9168-bcdec1051da4","owner":[],"postedDate":"July 16th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":34532807,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":34532808,"name":"Health sciences/Anatomy/Kidney"}],"tags":[],"updatedAt":"2024-11-11T16:00:12+00:00","versionOfRecord":{"articleIdentity":"rs-4613439","link":"https://doi.org/10.1038/s41598-024-77582-5","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2024-11-08 15:57:17","publishedOnDateReadable":"November 8th, 2024"},"versionCreatedAt":"2024-07-16 02:17:02","video":"","vorDoi":"10.1038/s41598-024-77582-5","vorDoiUrl":"https://doi.org/10.1038/s41598-024-77582-5","workflowStages":[]},"version":"v1","identity":"rs-4613439","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4613439","identity":"rs-4613439","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}