Enhancing Label-Free Fluorescence Lifetime Imaging for Intraoperative Tumor Margin Delineation in Head and Neck Cancer using Data-Centric AI

Enhancing Label-Free Fluorescence Lifetime Imaging for Intraoperative Tumor Margin Delineation in Head and Neck Cancer using Data-Centric AI | 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 Enhancing Label-Free Fluorescence Lifetime Imaging for Intraoperative Tumor Margin Delineation in Head and Neck Cancer using Data-Centric AI Mohamed Abul Hassan, Lisanne Kraft, Muhammad Adeel Azam, Julien Bec, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6673642/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 Precise intraoperative delineation of tumor margin is critical for maximizing resection completeness and minimizing recurrence in head and neck cancers (HNC). Label-free Fluorescence Lifetime Imaging (FLIm), which captures the fluorescence decay characteristics of endogenous molecules, offers a real-time method for differentiating malignant from healthy tissue during surgery without the need of contrast agents. Here, we present a data-centric artificial intelligence (AI) framework to enhance the robustness and accuracy of FLIm-based classification models for HNC. FLIm data were collected in vivo from 92 patients undergoing both transoral robotic surgery (TORS) and non-TORS procedures for HNC using a multispectral FLIm device with 355 nm excitation. To improve model performance, a data-centric approach leveraging confident learning was implemented to identify and exclude instances with low quality. An interpretability framework was further integrated to quantify feature contributions and elucidate FLIm-derived sources of contrast. Under a leave-one-patient-out cross-validation scheme, the model demonstrated a strong discriminative ability with an area under the receiver operating characteristic curve of 0.94 in differentiating healthy versus cancerous tissue. In tumor boundary regions, borderline predictions revealed transitional tissue properties, with strong correlations observed between model predictions and FLIm parameters in spectral channels corresponding to NADH and FAD—key metabolic cofactors indicative of cellular metabolic shifts at tumor margins. The impact of tumor anatomical site (base of tongue, palatine tonsil, oral tongue) and p16+ (HPV) status on classification performance was also assessed. These findings underscore the potential of label-free FLIm to provide accurate, real-time guidance for intraoperative margin assessment and dysplasia grading, advancing surgical precision in head and neck surgical oncology. Health sciences/Oncology/Surgical oncology Biological sciences/Cancer/Oral cancer Full Text Additional Declarations No competing interests reported. Supplementary Files ManuscriptDCHNCv3Supplementary.docx 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-6673642","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":459360070,"identity":"616d70a3-92e0-4cff-b003-e9c3fe34088c","order_by":0,"name":"Mohamed Abul Hassan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYFACHgYGxgYQI/kAAwMbGIEFidGSlkCylhwDqHoCWnTbzx7d8HGHXTQ/e843iR9ldol9/AcYH7xtw63F7Exe2s2ZZ5JzZ/a83SbZcy45sU0igdlwLj4tB3LMbvO2MeduuJG7TZqxjdmYTYKBTZoXn5bzb0Ba6nP338h5BtRSb8zGf4D9N14tN8C2HM7dIJHDBtRyWI6NIYGNGb+WN2Y3Z7Ydz51x5pmxZc+543JsEonNknPO4XNYjtmNj23Vuf3tyQ9v/Cir5pHvP3zww5sy3FqwAUg0jYJRMApGwSigAAAAjk1UDXD3m7YAAAAASUVORK5CYII=","orcid":"","institution":"University of California, Davis","correspondingAuthor":true,"prefix":"","firstName":"Mohamed","middleName":"Abul","lastName":"Hassan","suffix":""},{"id":459360071,"identity":"a820f003-0772-4a17-993a-e4c72eeb7f92","order_by":1,"name":"Lisanne Kraft","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Lisanne","middleName":"","lastName":"Kraft","suffix":""},{"id":459360072,"identity":"3eaae745-1c0e-4a78-8314-7166597e536b","order_by":2,"name":"Muhammad Adeel Azam","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"Adeel","lastName":"Azam","suffix":""},{"id":459360073,"identity":"6417a4e6-0040-40a7-bfec-7b8bf8da03e4","order_by":3,"name":"Julien Bec","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Julien","middleName":"","lastName":"Bec","suffix":""},{"id":459360074,"identity":"cc534b80-12b0-4336-b27b-1a2bd4397da2","order_by":4,"name":"Kelsey T Hadfield","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Kelsey","middleName":"T","lastName":"Hadfield","suffix":""},{"id":459360075,"identity":"74f48a2f-fb0f-4d50-8df8-f67a6409d2bb","order_by":5,"name":"Jinyi Qi","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Jinyi","middleName":"","lastName":"Qi","suffix":""},{"id":459360076,"identity":"6b6ca1e4-e865-42f7-9ea5-1d842de1ccee","order_by":6,"name":"Dorina Gui","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Dorina","middleName":"","lastName":"Gui","suffix":""},{"id":459360077,"identity":"e50b54fb-2945-4ce0-848c-7f52c3e9e8f0","order_by":7,"name":"Arnaud Bewley","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Arnaud","middleName":"","lastName":"Bewley","suffix":""},{"id":459360078,"identity":"9c992607-f9aa-46b4-9a35-2bac8e04a226","order_by":8,"name":"Marianne Abouyared","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Marianne","middleName":"","lastName":"Abouyared","suffix":""},{"id":459360079,"identity":"adeb9cba-0933-4092-bf78-96c9df140fb8","order_by":9,"name":"Jonathan Sorger","email":"","orcid":"","institution":"Intuitive Surgical (United States)","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Sorger","suffix":""},{"id":459360080,"identity":"1f0688ba-045b-4017-8eb4-64ac3464668b","order_by":10,"name":"Gregory Farwell","email":"","orcid":"","institution":"University of Pennsylvania","correspondingAuthor":false,"prefix":"","firstName":"Gregory","middleName":"","lastName":"Farwell","suffix":""},{"id":459360081,"identity":"c30aef7a-3464-43e5-8b61-936861b14a7a","order_by":11,"name":"Andrew C. Birkeland","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"C.","lastName":"Birkeland","suffix":""},{"id":459360082,"identity":"2ccde657-c198-4c66-8bdd-6c0739a6fe5c","order_by":12,"name":"Laura Marcu","email":"","orcid":"","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"","lastName":"Marcu","suffix":""}],"badges":[],"createdAt":"2025-05-15 14:38:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6673642/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6673642/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83321224,"identity":"0c351d3d-dbeb-4704-8d32-2b1518650487","added_by":"auto","created_at":"2025-05-23 03:54:38","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1636767,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptDCHNCv43.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6673642/v1_covered_7d440e71-d06a-41dc-bd2e-59e599225c27.pdf"},{"id":83320758,"identity":"ddafd02c-80c7-452c-acc5-9213b87898cd","added_by":"auto","created_at":"2025-05-23 03:46:35","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1094722,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptDCHNCv3Supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-6673642/v1/ad7684788accc6dbafbb467f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eEnhancing Label-Free Fluorescence Lifetime Imaging for Intraoperative Tumor Margin Delineation in Head and Neck Cancer using Data-Centric AI\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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-6673642/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6673642/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Precise intraoperative delineation of tumor margin is critical for maximizing resection completeness and minimizing recurrence in head and neck cancers (HNC). Label-free Fluorescence Lifetime Imaging (FLIm), which captures the fluorescence decay characteristics of endogenous molecules, offers a real-time method for differentiating malignant from healthy tissue during surgery without the need of contrast agents. Here, we present a data-centric artificial intelligence (AI) framework to enhance the robustness and accuracy of FLIm-based classification models for HNC. FLIm data were collected in vivo from 92 patients undergoing both transoral robotic surgery (TORS) and non-TORS procedures for HNC using a multispectral FLIm device with 355 nm excitation. To improve model performance, a data-centric approach leveraging confident learning was implemented to identify and exclude instances with low quality. An interpretability framework was further integrated to quantify feature contributions and elucidate FLIm-derived sources of contrast. Under a leave-one-patient-out cross-validation scheme, the model demonstrated a strong discriminative ability with an area under the receiver operating characteristic curve of 0.94 in differentiating healthy versus cancerous tissue. In tumor boundary regions, borderline predictions revealed transitional tissue properties, with strong correlations observed between model predictions and FLIm parameters in spectral channels corresponding to NADH and FAD—key metabolic cofactors indicative of cellular metabolic shifts at tumor margins. The impact of tumor anatomical site (base of tongue, palatine tonsil, oral tongue) and p16+ (HPV) status on classification performance was also assessed. These findings underscore the potential of label-free FLIm to provide accurate, real-time guidance for intraoperative margin assessment and dysplasia grading, advancing surgical precision in head and neck surgical oncology.","manuscriptTitle":"Enhancing Label-Free Fluorescence Lifetime Imaging for Intraoperative Tumor Margin Delineation in Head and Neck Cancer using Data-Centric AI","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-23 03:46:30","doi":"10.21203/rs.3.rs-6673642/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":"adf25777-4887-488f-8972-31f868447861","owner":[],"postedDate":"May 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48938583,"name":"Health sciences/Oncology/Surgical oncology"},{"id":48938584,"name":"Biological sciences/Cancer/Oral cancer"}],"tags":[],"updatedAt":"2025-05-23T03:46:30+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-23 03:46:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6673642","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6673642","identity":"rs-6673642","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}