A new class of dynamic contrast-enhanced MRI features for breast lesion classification

A new class of dynamic contrast-enhanced MRI features for breast lesion classification | 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 A new class of dynamic contrast-enhanced MRI features for breast lesion classification Gianluca Morcaldi, Maria Evelina Fantacci, Claudio Gasperi, Chiara Iacconi, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7740979/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 We propose an ensemble learning approach to classify malignant versus benign breast lesions leveraging morphological and dynamic features derived from Magnetic Resonance Images (MRI). The analysis has been performed on 164 breast lesions of the publicly available “Advanced MRI Breast Lesions” dataset from The Cancer Imaging Archive, containing T2-weighted and Dynamic Contrast-Enhanced (DCE)-MRI sequences, along with the segmentation masks of suspicious lesions. After extracting radiomic features using Pyradiomics Python package, we computed dynamic features from DCE-MRI kinetic curves, which describe the contrast agent wash-in and wash-out. These features have been defined as the derivatives of image intensity measures, like mean and standard deviation, computed inside the lesion masks on the 5 DCE-MRI time steps. We trained and evaluated an eXtreme Gradient Boosting (XGBoost) classifier, experimenting with different feature combinations in a stratified 5-fold cross-validation scheme. The best model trained on T2-weighted MRI morphological features achieved an Area Under the Curve (AUC) score of 0.83±0.04 on the independent test set consisting of 20 lesions, while the model using only dynamic features performed an AUC of 0.91±0.03. Despite being obtained on a limited size test sample, these results show the great potential of features derived from DCE images in breast lesions classification. Biological sciences/Cancer Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Health sciences/Medical research Health sciences/Oncology Full Text Additional Declarations No competing interests reported. Supplementary Files Supplementary.pdf 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-7740979","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":531674391,"identity":"07d2db32-d258-4b5c-a174-8b21c8a57d15","order_by":0,"name":"Gianluca Morcaldi","email":"data:image/png;base64,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","orcid":"","institution":"University of Pisa","correspondingAuthor":true,"prefix":"","firstName":"Gianluca","middleName":"","lastName":"Morcaldi","suffix":""},{"id":531674392,"identity":"f0815738-bea2-4946-8708-d15e698fd2a2","order_by":1,"name":"Maria Evelina Fantacci","email":"","orcid":"","institution":"University of Pisa","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Evelina","lastName":"Fantacci","suffix":""},{"id":531674393,"identity":"68d0c7e0-8ad6-4f2b-851e-471ba3b8c93d","order_by":2,"name":"Claudio Gasperi","email":"","orcid":"","institution":"University of Pisa","correspondingAuthor":false,"prefix":"","firstName":"Claudio","middleName":"","lastName":"Gasperi","suffix":""},{"id":531674394,"identity":"40b84178-b94b-40d8-ad9e-21644df765a7","order_by":3,"name":"Chiara Iacconi","email":"","orcid":"","institution":"Azienda USL Toscana Nord-Ovest","correspondingAuthor":false,"prefix":"","firstName":"Chiara","middleName":"","lastName":"Iacconi","suffix":""},{"id":531674395,"identity":"bdb71922-03d3-4426-a5dd-31e5b686c771","order_by":4,"name":"Elena Lorenzini","email":"","orcid":"","institution":"Azienda USL Toscana Nord-Ovest","correspondingAuthor":false,"prefix":"","firstName":"Elena","middleName":"","lastName":"Lorenzini","suffix":""},{"id":531674396,"identity":"1fb8e9a3-9d45-48dd-9e90-6732777bd23f","order_by":5,"name":"Emanuele Neri","email":"","orcid":"","institution":"University of Pisa","correspondingAuthor":false,"prefix":"","firstName":"Emanuele","middleName":"","lastName":"Neri","suffix":""},{"id":531674397,"identity":"3786f395-5e6b-4ac6-a877-33d426165c5c","order_by":6,"name":"Giuseppe Prencipe","email":"","orcid":"","institution":"University of Pisa","correspondingAuthor":false,"prefix":"","firstName":"Giuseppe","middleName":"","lastName":"Prencipe","suffix":""},{"id":531674398,"identity":"253359cb-eb3a-47ec-bdca-b848d6b13e76","order_by":7,"name":"Alessandra Retico","email":"","orcid":"","institution":"National Institute for Nuclear Physics","correspondingAuthor":false,"prefix":"","firstName":"Alessandra","middleName":"","lastName":"Retico","suffix":""},{"id":531674399,"identity":"4ceebda7-acad-4993-8b98-b92b14056f1a","order_by":8,"name":"Camilla Scapicchio","email":"","orcid":"","institution":"National Institute for Nuclear Physics","correspondingAuthor":false,"prefix":"","firstName":"Camilla","middleName":"","lastName":"Scapicchio","suffix":""},{"id":531674400,"identity":"a3e801fd-d24f-4e20-ab4c-20a5a61c4ebd","order_by":9,"name":"Maria Irene Tenerani","email":"","orcid":"","institution":"University of Pisa","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"Irene","lastName":"Tenerani","suffix":""},{"id":531674401,"identity":"5e49f56c-798f-474b-9c9c-5b445bc229e1","order_by":10,"name":"Francesca Lizzi","email":"","orcid":"","institution":"National Institute for Nuclear Physics","correspondingAuthor":false,"prefix":"","firstName":"Francesca","middleName":"","lastName":"Lizzi","suffix":""}],"badges":[],"createdAt":"2025-09-29 10:23:55","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7740979/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7740979/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94012830,"identity":"9db5fbbf-800a-40af-9cad-69670693b2cc","added_by":"auto","created_at":"2025-10-21 10:24:09","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11223,"visible":true,"origin":"","legend":"","description":"","filename":"917c067ee9ea426ba1fab48aee17952d.json","url":"https://assets-eu.researchsquare.com/files/rs-7740979/v1/554857835dfeabc7ad1969ab.json"},{"id":99308551,"identity":"1e63f41f-30f6-4b31-a62f-454f772059ea","added_by":"auto","created_at":"2025-12-31 16:08:45","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":615057,"visible":true,"origin":"","legend":"","description":"","filename":"AnewclassofdynamiccontrastenhancedMRIfeaturesforbreastlesionclassification.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7740979/v1_covered_b697eaee-ceec-44b6-a73a-5fdb7025b5f7.pdf"},{"id":94012829,"identity":"60377db9-0588-4b7a-a840-4f6f346d81fa","added_by":"auto","created_at":"2025-10-21 10:24:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":31135,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7740979/v1/d29a9748ff0c82025a0fc6bf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A new class of dynamic contrast-enhanced MRI features for breast lesion classification","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"","lastPublishedDoi":"10.21203/rs.3.rs-7740979/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7740979/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"We propose an ensemble learning approach to classify malignant versus benign breast lesions leveraging morphological and dynamic features derived from Magnetic Resonance Images (MRI). The analysis has been performed on 164 breast lesions of the publicly available “Advanced MRI Breast Lesions” dataset from The Cancer Imaging Archive, containing T2-weighted and Dynamic Contrast-Enhanced (DCE)-MRI sequences, along with the segmentation masks of suspicious lesions. After extracting radiomic features using Pyradiomics Python package, we computed dynamic features from DCE-MRI kinetic curves, which describe the contrast agent wash-in and wash-out. These features have been defined as the derivatives of image intensity measures, like mean and standard deviation, computed inside the lesion masks on the 5 DCE-MRI time steps. We trained and evaluated an eXtreme Gradient Boosting (XGBoost) classifier, experimenting with different feature combinations in a stratified 5-fold cross-validation scheme. The best model trained on T2-weighted MRI morphological features achieved an Area Under the Curve (AUC) score of 0.83±0.04 on the independent test set consisting of 20 lesions, while the model using only dynamic features performed an AUC of 0.91±0.03. Despite being obtained on a limited size test sample, these results show the great potential of features derived from DCE images in breast lesions classification.","manuscriptTitle":"A new class of dynamic contrast-enhanced MRI features for breast lesion classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-21 10:24:05","doi":"10.21203/rs.3.rs-7740979/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":"c5f8f656-f016-4a03-88ba-b1a178ad4685","owner":[],"postedDate":"October 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56525532,"name":"Biological sciences/Cancer"},{"id":56525533,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56525534,"name":"Physical sciences/Mathematics and computing"},{"id":56525535,"name":"Health sciences/Medical research"},{"id":56525536,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2025-12-23T04:09:26+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-21 10:24:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7740979","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7740979","identity":"rs-7740979","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}