SAM-KDNet: A Segmentation and Knowledge Distillation Framework for Automated CVM Stages Classification from CBCT

SAM-KDNet: A Segmentation and Knowledge Distillation Framework for Automated CVM Stages Classification from CBCT | 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 SAM-KDNet: A Segmentation and Knowledge Distillation Framework for Automated CVM Stages Classification from CBCT Omid Halimi Milani, Amanda Nikho, Lauren Mills, Marouane Tliba, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7991912/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 Objective . Automated skeletal maturity assessment is essential in orthodontic diagnosis and treatment planning, guiding the optimal timing of interventions such as myofunctional appliances and orthognathic surgery. This study aims to develop, test, and validate automated interpretable deep learning algorithms for the assessment and classification of cervical vertebrae maturation (CVM) stages from cone beam computed tomography (CBCT) scans. Methods. The sample consisted of 364 CBCT scans of orthodontic patients from private practices in the midwestern United States. The CVM stages were classified by two orthodontists and an oral and maxillofacial radiologist. A deep learning pipeline was designed that incorporated four innovations: (1) prompt-guided segmentation using SAM 1 for anatomical region localization, (2) demographic-aware modeling with age embeddings and gender conditioning, (3) knowledge distillation from a teacher network pretrained on a larger skeletal age dataset, and (4) tri-modal input fusion to emphasize clinically salient regions. Results. The best-performing model (88.54% accuracy, 5-fold cross-validation average) was the ResNet50 architecture with knowledge distillation, trained on a subset of 96 samples using our previously developed spheno-occipital synchondrosis staging model as the teacher network. Conclusion. Our findings demonstrate that combining anatomical cues, patient-specific information, and knowledge distillation enhances deep learning–based CVM staging and enables the models to effectively address the challenges of CBCT-based skeletal maturity assessment in orthodontics. This approach provides a foundation for more accurate and clinically meaningful integration of CBCT into orthodontic diagnostic workflows. Health sciences/Anatomy Biological sciences/Computational biology and bioinformatics Health sciences/Health care Physical sciences/Mathematics and computing Health sciences/Medical research Full Text Additional Declarations No competing interests reported. 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-7991912","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":549802590,"identity":"f125642d-016f-499c-99f8-634fd53d2d35","order_by":0,"name":"Omid Halimi Milani","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Omid","middleName":"Halimi","lastName":"Milani","suffix":""},{"id":549802592,"identity":"d482f92c-b5fc-4029-a1c6-89b0b79457ae","order_by":1,"name":"Amanda Nikho","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Amanda","middleName":"","lastName":"Nikho","suffix":""},{"id":549802593,"identity":"868a3483-2cfe-48af-aeaa-5e9b2b72419d","order_by":2,"name":"Lauren Mills","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Lauren","middleName":"","lastName":"Mills","suffix":""},{"id":549802594,"identity":"41862561-8f5e-41d4-a70d-bd93cec4b96d","order_by":3,"name":"Marouane Tliba","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Marouane","middleName":"","lastName":"Tliba","suffix":""},{"id":549802595,"identity":"293be63b-d457-4c70-9b11-090030eff30a","order_by":4,"name":"Veerasathpurush Allareddy","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Veerasathpurush","middleName":"","lastName":"Allareddy","suffix":""},{"id":549802596,"identity":"366d044e-69cc-4570-9dcb-1fd6d9736bfb","order_by":5,"name":"Rashid Ansari","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Rashid","middleName":"","lastName":"Ansari","suffix":""},{"id":549802597,"identity":"bde9ed06-4116-4e06-82fd-403c9abf2c7c","order_by":6,"name":"Ahmet Enis Cetin","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Ahmet","middleName":"Enis","lastName":"Cetin","suffix":""},{"id":549802598,"identity":"dc663abc-e394-416e-9398-160064431031","order_by":7,"name":"Ahmet Atakan","email":"","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":false,"prefix":"","firstName":"Ahmet","middleName":"","lastName":"Atakan","suffix":""},{"id":549802601,"identity":"0f5a1fdc-4f36-41bc-b611-9ba23f6008bc","order_by":8,"name":"Mohammed H. Elnagar","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwUlEQVRIiWNgGAWjYFAC5gYwxc/ACGIwE6OFEaJFsoFkLQYHIHYS1qDb3tj24Mcvu3zjG8mNHxgqrBMbCGkxO3Ow3bC3L9ly243EZgmGM+lEaLmR2CbB28NsAGIwMLYdJkLL/Ydtkn976g2MZ4C0/CNGyw3GNmmeH4cNDCRAWhqI0XImsU1atuG4gcSZh80SCcfSjQlrOX74mOSbP9UG/O3pDz98qLGWJagFDBjboIwEopSDwR/ilY6CUTAKRsEIBADY1EIbUHhJlQAAAABJRU5ErkJggg==","orcid":"","institution":"University of Illinois Chicago","correspondingAuthor":true,"prefix":"","firstName":"Mohammed","middleName":"H.","lastName":"Elnagar","suffix":""}],"badges":[],"createdAt":"2025-10-30 18:08:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7991912/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7991912/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96686422,"identity":"d7f8607e-1d77-4780-8a9f-cbde55f7c1dc","added_by":"auto","created_at":"2025-11-25 05:31:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8004628,"visible":true,"origin":"","legend":"","description":"","filename":"Binder1.1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7991912/v1/7c2abcb803187f3f7e22da32.pdf"},{"id":96686421,"identity":"12affac2-c051-4867-83bd-228f19842cdf","added_by":"auto","created_at":"2025-11-25 05:31:02","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10554,"visible":true,"origin":"","legend":"","description":"","filename":"b0245b3e5d594f2b8656f487acb93d48.json","url":"https://assets-eu.researchsquare.com/files/rs-7991912/v1/b879196bf9d5243c5e9ca14e.json"},{"id":106404638,"identity":"ea3f284e-efa8-41e6-945b-a8c64ee50944","added_by":"auto","created_at":"2026-04-08 09:16:25","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2389858,"visible":true,"origin":"","legend":"","description":"","filename":"Binder1.1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7991912/v1_covered_bb1f3cb6-a4a5-4a38-a23b-7de2e1c19c6a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"SAM-KDNet: A Segmentation and Knowledge Distillation Framework for Automated CVM Stages Classification from CBCT","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-7991912/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7991912/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eObjective\u003c/strong\u003e. Automated skeletal maturity assessment is essential in orthodontic diagnosis and treatment planning, guiding the optimal timing of interventions such as myofunctional appliances and orthognathic surgery. This study aims to develop, test, and validate automated interpretable deep learning algorithms for the assessment and classification of cervical vertebrae maturation (CVM) stages from cone beam computed tomography (CBCT) scans.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods.\u003c/strong\u003e The sample consisted of 364 CBCT scans of orthodontic patients from private practices in the midwestern United States. The CVM stages were classified by two orthodontists and an oral and maxillofacial radiologist. A deep learning pipeline was designed that incorporated four innovations: (1) prompt-guided segmentation using SAM 1 for anatomical region localization, (2) demographic-aware modeling with age embeddings and gender conditioning, (3) knowledge distillation from a teacher network pretrained on a larger skeletal age dataset, and (4) tri-modal input fusion to emphasize clinically salient regions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults.\u003c/strong\u003e The best-performing model (88.54% accuracy, 5-fold cross-validation average) was the ResNet50 architecture with knowledge distillation, trained on a subset of 96 samples using our previously developed spheno-occipital synchondrosis staging model as the teacher network.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion.\u003c/strong\u003e Our findings demonstrate that combining anatomical cues, patient-specific information, and knowledge distillation enhances deep learning–based CVM staging and enables the models to effectively address the challenges of CBCT-based skeletal maturity assessment in orthodontics. This approach provides a foundation for more accurate and clinically meaningful integration of CBCT into orthodontic diagnostic workflows.\u003c/p\u003e","manuscriptTitle":"SAM-KDNet: A Segmentation and Knowledge Distillation Framework for Automated CVM Stages Classification from CBCT","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-25 05:30:58","doi":"10.21203/rs.3.rs-7991912/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":"252bb318-c732-4f77-bc6b-ff0c1d7e3a22","owner":[],"postedDate":"November 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":58495293,"name":"Health sciences/Anatomy"},{"id":58495294,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":58495295,"name":"Health sciences/Health care"},{"id":58495296,"name":"Physical sciences/Mathematics and computing"},{"id":58495297,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-04-07T05:56:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-25 05:30:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7991912","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7991912","identity":"rs-7991912","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}