From Qubits to Cavities: Quantum-Inspired U-Net for Dental Caries Segmentation

From Qubits to Cavities: Quantum-Inspired U-Net for Dental Caries Segmentation | 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 From Qubits to Cavities: Quantum-Inspired U-Net for Dental Caries Segmentation Mandeep Kaur, Amar Nath, Utkarsh Niranjan This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9257344/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Objectives: The objective of this study is to develop a hybrid quantum-classical deep learning framework for accurate dental caries detection from radiographs, and to evaluate its performance by conducting segmentation analysis at the pixel level to help clinicians make informed decisions in a variety of dental practices. Materials and Methods: A large dataset of 3200 high-quality intraoral images, annotated by dental experts, was used for training and validation. The feature extraction was based on a U-Net architecture, and a quantum-enhanced latent layer was implemented on a parameterized quantum circuit. Dice loss combined with a binary cross-entropy loss function was adopted for training. The developed model was evaluated on an unseen test set using several metrics. Results: In experimental results, the hybrid deep learning-based model achieved high accuracy with 0.93 ± 0.02 in Dice coeffcient and 0.87 ± 0.02 in IoU. The corresponding precision and recall are 0.94±0.01 and 0.91±0.02, respectively. The improvements of the proposed model over the classical U-Net are statistically significant (p < 0.01). The proposed model also shows consistent performance for different categories of lesions, including deep and early stages of caries. Conclusions: Quantum-enhanced feature processing applied to a U-Net-based model significantly improves accuracy and robustness for dental radiographic image segmentation. It enables precise caries detection and reduces specialists’ diagnostic uncertainty. Clinical Relevance: This method offers a reliable aid for early caries detection and treatment planning, and can increase diagnostic consistency; thus supporting minimally invasive dentistry. Dental Caries Hybrid Quantum-classical Deep Learning Image Segmentation U-Net Dental Radiography Quantum Neural Networks Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 13 May, 2026 Reviewers agreed at journal 19 Apr, 2026 Reviewers agreed at journal 18 Apr, 2026 Reviewers invited by journal 18 Apr, 2026 Editor assigned by journal 01 Apr, 2026 Submission checks completed at journal 01 Apr, 2026 First submitted to journal 29 Mar, 2026 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-9257344","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":628817353,"identity":"51e7281c-cecb-41b4-8d6b-b27330ae5ee5","order_by":0,"name":"Mandeep Kaur","email":"","orcid":"","institution":"Sant Longowal Institute of Engineering and Technology","correspondingAuthor":false,"prefix":"","firstName":"Mandeep","middleName":"","lastName":"Kaur","suffix":""},{"id":628817354,"identity":"66486a43-4dd7-406f-af07-4f6cecc4c078","order_by":1,"name":"Amar Nath","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4klEQVRIie3PsQrCMBSF4RsCTpWudWlfIRJwUnyVBEEXEaGrQ1zaRcwmAV9CKDi3BFxdC1mcnBVBHESsuIm0dXPIN2U4P+ECWNYfYy5gkb7fSNQpCGsJVCTsl4SkryWrsXVX+niZzu4Tus/m6fkGgStwdChLPDOka7UjYSfnIlMM2ipFMSn9xjCKnQbh2xwJ7TBAG0CRV1YEZnTBzoPwRGZC3xn0KxNixhQ3I8I3wIUuzueVSduMQ9xcUq5etyyG3kDpisQ3owQ7V59LqfXp1u32ZBwfS5NPxRj/srcsy7K+egIcIEup91C2ZwAAAABJRU5ErkJggg==","orcid":"","institution":"Sant Longowal Institute of Engineering and Technology","correspondingAuthor":true,"prefix":"","firstName":"Amar","middleName":"","lastName":"Nath","suffix":""},{"id":628817364,"identity":"86497211-e09c-4e68-a80f-bd4cbc961059","order_by":2,"name":"Utkarsh Niranjan","email":"","orcid":"","institution":"Sant Longowal Institute of Engineering and Technology","correspondingAuthor":false,"prefix":"","firstName":"Utkarsh","middleName":"","lastName":"Niranjan","suffix":""}],"badges":[],"createdAt":"2026-03-29 08:23:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9257344/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9257344/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108006103,"identity":"d2204ea9-bd8a-4dd3-a62a-e499aede9a15","added_by":"auto","created_at":"2026-04-28 12:53:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":581590,"visible":true,"origin":"","legend":"","description":"","filename":"SpringerClinicalOralInvestigationMandeep.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9257344/v1_covered_8930b03c-9d4d-4a69-8d42-023cd4b8a8f9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Qubits to Cavities: Quantum-Inspired U-Net for Dental Caries Segmentation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"clinical-oral-investigations","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cloi","sideBox":"Learn more about [Clinical Oral Investigations](http://link.springer.com/journal/784)","snPcode":"784","submissionUrl":"https://submission.nature.com/new-submission/784/3","title":"Clinical Oral Investigations","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Dental Caries, Hybrid Quantum-classical Deep Learning, Image Segmentation, U-Net, Dental Radiography, Quantum Neural Networks","lastPublishedDoi":"10.21203/rs.3.rs-9257344/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9257344/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eObjectives: The objective of this study is to develop a hybrid quantum-classical deep learning framework for accurate dental caries detection from radiographs, and to evaluate its performance by conducting segmentation analysis at the pixel level to help clinicians make informed decisions in a variety of dental practices.\u003c/p\u003e\n\u003cp\u003eMaterials and Methods: A large dataset of 3200 high-quality intraoral images, annotated by dental experts, was used for training and validation. The feature extraction was based on a U-Net architecture, and a quantum-enhanced latent layer was implemented on a parameterized quantum circuit. Dice loss combined with a binary cross-entropy loss function was adopted for training. The developed model was evaluated on an unseen test set using several metrics.\u003c/p\u003e\n\u003cp\u003eResults: In experimental results, the hybrid deep learning-based model achieved high accuracy with 0.93 ± 0.02 in Dice coeffcient and 0.87 ± 0.02 in IoU. The corresponding precision and recall are 0.94±0.01 and 0.91±0.02, respectively. The improvements of the proposed model over the classical U-Net are statistically significant (p \u0026lt; 0.01). The proposed model also shows consistent performance for different categories of lesions, including deep and early stages of caries.\u003c/p\u003e\n\u003cp\u003eConclusions: Quantum-enhanced feature processing applied to a U-Net-based model significantly improves accuracy and robustness for dental radiographic image segmentation. It enables precise caries detection and reduces specialists’ diagnostic uncertainty.\u003c/p\u003e\n\u003cp\u003eClinical Relevance: This method offers a reliable aid for early caries detection and treatment planning, and can increase diagnostic consistency; thus supporting minimally invasive dentistry.\u003c/p\u003e","manuscriptTitle":"From Qubits to Cavities: Quantum-Inspired U-Net for Dental Caries Segmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-27 09:35:44","doi":"10.21203/rs.3.rs-9257344/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-14T01:37:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"270197022703045659371495712661054485815","date":"2026-04-19T19:32:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"192588794272189360933978888532922767267","date":"2026-04-19T00:48:01+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-18T22:55:35+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-01T07:11:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-01T07:11:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Clinical Oral Investigations","date":"2026-03-29T08:11:56+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"clinical-oral-investigations","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cloi","sideBox":"Learn more about [Clinical Oral Investigations](http://link.springer.com/journal/784)","snPcode":"784","submissionUrl":"https://submission.nature.com/new-submission/784/3","title":"Clinical Oral Investigations","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"373f957a-48c9-4216-8ed8-be82326f26d9","owner":[],"postedDate":"April 27th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-14T01:37:25+00:00","index":16,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T09:35:45+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-27 09:35:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9257344","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9257344","identity":"rs-9257344","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}