Deep Learning-Based 3D and 2D Approaches for Skeletal Muscle Segmentation on Low-Dose CT Images

Deep Learning-Based 3D and 2D Approaches for Skeletal Muscle Segmentation on Low-Dose CT 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 Deep Learning-Based 3D and 2D Approaches for Skeletal Muscle Segmentation on Low-Dose CT Images Giuseppe Timpano, Pierangelo Veltri, Patrizia Vizza, Giuseppe Lucio, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6786492/v2 This work is licensed under a CC BY 4.0 License Status: Posted Version 2 posted You are reading this latest preprint version Show more versions Abstract Automated segmentation of skeletal muscle from computed tomography (CT) images is essential for large-scale quantitative body composition analysis. However, manual segmentation is time-consuming and impractical for routine or high-throughput use. This study presents a systematic comparison of two-dimensional (2D) and three-dimensional (3D) deep learning architectures for segmenting skeletal muscle at the anatomically standardized level of the third lumbar vertebra (L3) in low-dose computed tomography (LDCT) scans. We implemented and evaluated the DeepLabv3+ (2D) and UNet3+ (3D) architectures on a curated dataset of 534 LDCT scans, applying preprocessing protocols, L3 slice selection, and region of interest extraction. The model performance was evaluated using a comprehensive set of evaluation metrics, including Dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (HD95). DeepLabv3+ achieved the highest segmentation accuracy (DSC = 0.982 ± 0.010, HD95 = 1.04 ± 0.46 mm), while UNet3+ showed competitive performance (DSC = 0.967 ± 0.013, HD95 = 1.27 ± 0.58 mm) with 26 times fewer parameters (1.27 million vs. 33.6 million) and lower inference time. Both models exceeded or matched results reported in the recent CT-based muscle segmentation literature. This work offers practical insights into architecture selection for automated LDCT-based muscle segmentation workflows, with a focus on the L3 vertebral level, which remains the gold standard in muscle quantification protocols. Biomedical Engineering Artificial Intelligence and Machine Learning Skeletal muscle Deep Learning LDCT Segmentation UNet DeepLab Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 2 posted You are reading this latest preprint version Show more versions 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-6786492","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":465626340,"identity":"fcebbd6c-c8aa-4cc6-ac52-79ffb72531f4","order_by":0,"name":"Giuseppe Timpano","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABD0lEQVRIiWNgGAWjYDACZjiLsQFM8QPxASDmIayFDapFsgGqBbceuBYobXAAysClRd6dx/jDxx135Bjkm1s3F9TckTO+kXvwwI8/DDL2OLQYHuYxk5x55pkx0GFtt2cce2ZsdiMv4WAPD26HGTbzmDHzth1ObABp4WE7nLjtRo7BYQYJvFqMP/9tO1wP0fLvcOLmGSAtBri1yDPzGEgzth1OADsMZN0GCZCWBNxaDJjZyiR7zzwzbGNLbLs9s++wscSZNwYHew5I8PAcwGFL/+HNH37uuCPPz3z82e2Cb4fl+NtzjD/8+GNjz96AwxaQUYwNB8DRwowkIYHDWUBbGqBaGNC0jIJRMApGwSiAAwAsD1mmmQOCfwAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0009-0001-6417-6561","institution":"Department of Surgical and Medical Sciences, Magna Graecia University","correspondingAuthor":true,"prefix":"","firstName":"Giuseppe","middleName":"","lastName":"Timpano","suffix":""},{"id":465626341,"identity":"2ebb7dd3-8e87-49dc-9145-fc88f8306d3d","order_by":1,"name":"Pierangelo Veltri","email":"","orcid":"https://orcid.org/0000-0003-2494-0294","institution":"DIMES, University of Calabria","correspondingAuthor":false,"prefix":"","firstName":"Pierangelo","middleName":"","lastName":"Veltri","suffix":""},{"id":465626342,"identity":"d079c7ef-4934-4bd5-ab98-baadc863d695","order_by":2,"name":"Patrizia Vizza","email":"","orcid":"https://orcid.org/0000-0002-0014-478X","institution":"Department of Surgical and Medical Sciences, Magna Graecia University","correspondingAuthor":false,"prefix":"","firstName":"Patrizia","middleName":"","lastName":"Vizza","suffix":""},{"id":487736561,"identity":"c0046276-29c0-401d-8a21-e8e87440e9d4","order_by":3,"name":"Giuseppe Lucio","email":"","orcid":"https://orcid.org/0000-0003-0332-2606","institution":"Cascini","correspondingAuthor":false,"prefix":"","firstName":"Giuseppe","middleName":"","lastName":"Lucio","suffix":""},{"id":487736562,"identity":"42308d76-b3f3-4f2e-916f-0ff11b1e3b62","order_by":4,"name":"Francesco","email":"","orcid":"","institution":"Manti","correspondingAuthor":false,"prefix":"","firstName":"","middleName":"","lastName":"Francesco","suffix":""}],"badges":[],"createdAt":"2025-05-30 16:45:54","currentVersionCode":2,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6786492/v2","doiUrl":"https://doi.org/10.21203/rs.3.rs-6786492/v2","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88135851,"identity":"328e6733-1bc6-4e9f-b1de-186628b215f9","added_by":"auto","created_at":"2025-08-01 21:45:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":669844,"visible":true,"origin":"","legend":"","description":"","filename":"Sarcopenia2025.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6786492/v2_covered_286cfa6a-127e-4aeb-89c1-2a3f3e883391.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eDeep Learning-Based 3D and 2D Approaches for Skeletal Muscle Segmentation on Low-Dose CT Images\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Magna Graecia University","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":"Skeletal muscle, Deep Learning, LDCT Segmentation, UNet, DeepLab","lastPublishedDoi":"10.21203/rs.3.rs-6786492/v2","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6786492/v2","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAutomated segmentation of skeletal muscle from computed tomography (CT) images is essential for large-scale quantitative body composition analysis. However, manual segmentation is time-consuming and impractical for routine or high-throughput use. This study presents a systematic comparison of two-dimensional (2D) and three-dimensional (3D) deep learning architectures for segmenting skeletal muscle at the anatomically standardized level of the third lumbar vertebra (L3) in low-dose computed tomography (LDCT) scans.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eWe implemented and evaluated the DeepLabv3+ (2D) and UNet3+ (3D) architectures on a curated dataset of 534 LDCT scans, applying preprocessing protocols, L3 slice selection, and region of interest extraction. The model performance was evaluated using a comprehensive set of evaluation metrics, including Dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (HD95). DeepLabv3+ achieved the highest segmentation accuracy (DSC = 0.982 ± 0.010, HD95 = 1.04 ± 0.46 mm), while UNet3+ showed competitive performance (DSC = 0.967 ± 0.013, HD95 = 1.27 ± 0.58 mm) with 26 times fewer parameters (1.27 million vs. 33.6 million) and lower inference time.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eBoth models exceeded or matched results reported in the recent CT-based muscle segmentation literature. This work offers practical insights into architecture selection for automated LDCT-based muscle segmentation workflows, with a focus on the L3 vertebral level, which remains the gold standard in muscle quantification protocols.\u003c/p\u003e","manuscriptTitle":"Deep Learning-Based 3D and 2D Approaches for Skeletal Muscle Segmentation on Low-Dose CT Images","msid":"","msnumber":"","nonDraftVersions":[{"code":2,"date":"2025-07-31 18:16:27","doi":"10.21203/rs.3.rs-6786492/v2","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}},{"code":1,"date":"2025-06-03 05:40:23","doi":"10.21203/rs.3.rs-6786492/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":"70b86cee-20d0-4517-ad5d-48e57de33412","owner":[],"postedDate":"July 31st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":52522370,"name":"Biomedical Engineering"},{"id":52522371,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2025-08-01T21:38:20+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-31 18:16:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v2","identity":"rs-6786492","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6786492","identity":"rs-6786492","version":["v2"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}