{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T23:05:17Z","timestamp":1768086317285,"version":"3.49.0"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T00:00:00Z","timestamp":1757289600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100006098","name":"Radiological Society of North America","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006098","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Head and neck squamous cell carcinomas (HNSCC) present a significant clinical challenge due to high recurrence rates despite advances in radiation and chemotherapy. Early detection of recurrence is critical for optimizing treatment outcomes and improving patient survival.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We developed two artificial intelligence (AI) pipelines\u2014(1) machine learning models trained on radiomic and clinical data and (2) a Vision Transformer-based model directly applied to imaging data\u2014to predict HNSCC recurrence using pre- and post-treatment PET\/CT scans from a cohort of 249 patients. We incorporated Test-Time Augmentation (TTA) and Conformal Prediction to quantify prediction uncertainty and enhance model reliability.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The machine learning models achieved an average AUC of 0.820. The vision transformer model showed moderate performance (AUC = 0.658). Uncertainty quantification enabled the exclusion of ambiguous predictions, improving accuracy among more confident cases.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Our machine learning models achieved strong performance in predicting HNSCC recurrence from radiomic and clinical features. Incorporating uncertainty quantification further improved predictive performance and reliability.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frai.2025.1623393","type":"journal-article","created":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T05:26:28Z","timestamp":1757309188000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Enhancing prediction of primary site recurrence in head and neck cancer using radiomics and uncertainty estimation"],"prefix":"10.3389","volume":"8","author":[{"given":"Yu","family":"Hu","sequence":"first","affiliation":[]},{"given":"Kimberly","family":"Taing","sequence":"additional","affiliation":[]},{"given":"Jing","family":"Wang","sequence":"additional","affiliation":[]},{"given":"David","family":"Sher","sequence":"additional","affiliation":[]},{"given":"Michael","family":"Dohopolski","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,9,8]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1186\/s43046-024-00222-6","article-title":"Predicting disease recurrence in breast cancer patients using machine learning models with clinical and radiomic characteristics: a retrospective study","volume":"36","author":"Azeroual","year":"2024","journal-title":"J. Egypt. Natl. Canc. Inst"},{"key":"B2","doi-asserted-by":"publisher","first-page":"942","DOI":"10.1080\/0284186X.2021.1925343","article-title":"The risk of recurrence in surgically treated head and neck squamous cell carcinomas: a conditional probability approach","volume":"60","author":"Borsetto","year":"2021","journal-title":"Acta Oncol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"458","DOI":"10.14639\/0392-100X-1246","article-title":"Prognostic factors in head and neck cancer: a 10-year retrospective analysis in a single-institution in Italy","volume":"37","author":"Cadoni","year":"2017","journal-title":"Acta Otorhinolaryngol. Ital"},{"key":"B4","doi-asserted-by":"publisher","first-page":"421","DOI":"10.1177\/0194599815621742","article-title":"Detecting residual\/recurrent head neck squamous cell carcinomas using pet or PET\/CT: systematic review and meta-analysis","volume":"154","author":"Cheung","year":"2016","journal-title":"Otolaryngol. Head Neck Surg"},{"key":"B5","doi-asserted-by":"publisher","first-page":"1384","DOI":"10.3390\/diagnostics11081384","article-title":"Transmed: Transformers advance multi-modal medical image classification","volume":"11","author":"Dai","year":"2021","journal-title":"Diagnostics"},{"key":"B6","doi-asserted-by":"publisher","first-page":"4232","DOI":"10.3390\/cancers15174232","article-title":"Radiotherapy and systemic therapies: Focus on head and neck cancer","volume":"15","author":"De Felice","year":"2023","journal-title":"Cancers"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929","article-title":"An image is worth 16x16 words: transformers for image recognition at scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv [preprint"},{"key":"B8","doi-asserted-by":"publisher","first-page":"607","DOI":"10.1186\/s12885-024-12331-5","article-title":"Reference-free inferring of transcriptomic events in cancer cells on single-cell data","volume":"24","author":"Eralp","year":"2024","journal-title":"BMC Cancer"},{"key":"B9","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1016\/j.ctro.2021.03.002","article-title":"Ultrasound delta-radiomics during radiotherapy to predict recurrence in patients with head and neck squamous cell carcinoma","volume":"28","author":"Fatima","year":"2021","journal-title":"Clin. Transl. Radiat. Oncol"},{"key":"B10","doi-asserted-by":"publisher","first-page":"20200073","DOI":"10.1259\/bjro.20200073","article-title":"Radiomics ai prediction for head and neck squamous cell carcinoma (HNSCC) prognosis and recurrence with target volume approach","volume":"3","author":"Fh","year":"2021","journal-title":"BJR Open"},{"key":"B11","doi-asserted-by":"publisher","first-page":"e0277168","DOI":"10.1371\/journal.pone.0277168","article-title":"Utility of adding radiomics to clinical features in predicting the outcomes of radiotherapy for head and neck cancer using machine learning","volume":"17","author":"Gangil","year":"2022","journal-title":"PLoS ONE"},{"key":"B12","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1016\/j.ins.2018.07.076","article-title":"Aecid: asymmetric entropy for classifying imbalanced data","volume":"467","author":"Guermazi","year":"2018","journal-title":"Inf. Sci"},{"key":"B13","doi-asserted-by":"publisher","first-page":"S521","DOI":"10.1016\/j.jacr.2023.08.008","article-title":"ACR appropriateness criteria\u00ae staging and post-therapy assessment of head and neck cancer","volume":"20","author":"Gule-Monroe","year":"2020","journal-title":"J. Am. Coll. Radiol"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1738932","DOI":"10.1155\/2021\/1738932","article-title":"Epidemiological trends of head and neck cancer: a population-based study","volume":"2021","author":"Guo","year":"2021","journal-title":"Biomed Res. Int"},{"key":"B15","doi-asserted-by":"publisher","first-page":"2817","DOI":"10.1002\/cncr.34823","article-title":"Patterns of recurrence in head and neck squamous cell carcinoma to inform personalized surveillance protocols","volume":"129","author":"Haring","year":"2023","journal-title":"Cancer"},{"key":"B16","doi-asserted-by":"publisher","first-page":"6202","DOI":"10.1109\/WACV56688.2023.00614","article-title":"\u201cHiformer: hierarchical multi-scale representations using transformers for medical image segmentation,\u201d","author":"Heidari","year":"2023"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2973","DOI":"10.1038\/bjc.2013.668","article-title":"18F-FDG PET\/CT surveillance at 3-6 and 12 months for detection of recurrence and second primary cancer in patients with head and neck squamous cell carcinoma","volume":"109","author":"Kim","year":"2013","journal-title":"Br. J. Cancer"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1078","DOI":"10.3348\/kjr.2022.0299","article-title":"Development and validation of a model using radiomics features from an apparent diffusion coefficient map to diagnose local tumor recurrence in patients treated for head and neck squamous cell carcinoma","volume":"23","author":"Kim","year":"2022","journal-title":"Korean J. Radiol"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2505.18823","article-title":"Mslau-net: a hybird CNN-transformer network for medical image segmentation","author":"Lan","year":"2025","journal-title":"arXiv [preprint"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1487","DOI":"10.1001\/jamaoncol.2017.0973","article-title":"Patterns of treatment failure and postrecurrence outcomes among patients with locally advanced head and neck squamous cell carcinoma after chemoradiotherapy using modern radiation techniques","volume":"3","author":"Leeman","year":"2017","journal-title":"JAMA Oncol"},{"key":"B21","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1080\/03610920903140247","article-title":"Construction of an off-centered entropy for the supervised learning of imbalanced classes: some first results","volume":"39","author":"Lenca","year":"2010","journal-title":"Commun. Stat. Theory Methods"},{"key":"B22","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1007\/978-3-031-16852-9_2","article-title":"\u201cBenchmarking and boosting transformers for medical image classification,\u201d","author":"Ma","year":"2022"},{"key":"B23","doi-asserted-by":"publisher","first-page":"1444","DOI":"10.1056\/NEJMoa1514493","article-title":"PET-CT surveillance versus neck dissection in advanced head and neck cancer","volume":"374","author":"Mehanna","year":"2016","journal-title":"N. Engl. J. Med"},{"key":"B24","author":"Molnar","year":"2023"},{"key":"B25","doi-asserted-by":"publisher","first-page":"108879","DOI":"10.1016\/j.compbiomed.2024.108879","article-title":"Radiomics-based prediction of recurrence for head and neck cancer patients using data imbalanced correction","volume":"180","author":"Oka","year":"2024","journal-title":"Comput. Biol. Med"},{"key":"B26","first-page":"64","article-title":"\u201cNormalized nonconformity measures for regression conformal prediction,\u201d","author":"Papadopoulos","year":"2008"},{"key":"B27","doi-asserted-by":"publisher","first-page":"873","DOI":"10.6004\/jnccn.2020.0031","article-title":"Head and neck cancers, version 2.2020, NCCN clinical practice guidelines in oncology","volume":"18","author":"Pfister","year":"2020","journal-title":"J. Natl. Compr. Canc. Netw"},{"key":"B28","doi-asserted-by":"publisher","first-page":"2206","DOI":"10.1002\/lary.31188","article-title":"Higher rates of inadequate adjuvant radiation dose among older adults with head and neck cancer","volume":"134","author":"Raab","year":"2024","journal-title":"Laryngoscope"},{"key":"B29","doi-asserted-by":"publisher","first-page":"629","DOI":"10.2214\/AJR.14.14166","article-title":"Diagnostic accuracy of follow-up FDG pet or PET\/CT in patients with head and neck cancer after definitive treatment: a systematic review and meta-analysis","volume":"205","author":"Sheikhbahaei","year":"2015","journal-title":"Am. J. Roentgenol"},{"key":"B30","doi-asserted-by":"publisher","first-page":"264","DOI":"10.1245\/s10434-014-3951-8","article-title":"Cancer-specific mortality and competing mortality in patients with head and neck squamous cell carcinoma: a competing risk analysis","volume":"22","author":"Shen","year":"2015","journal-title":"Ann. Surg. Oncol"},{"key":"B31","doi-asserted-by":"publisher","first-page":"974467","DOI":"10.3389\/fonc.2022.974467","article-title":"Improving radiomic model reliability using robust features from perturbations for head-and-neck carcinoma","volume":"12","author":"Teng","year":"2022","journal-title":"Front. Oncol"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1174","DOI":"10.3390\/cancers15041174","article-title":"Radiomics applications in head and neck tumor imaging: a narrative review","volume":"15","author":"Tortora","year":"2023","journal-title":"Cancers"},{"key":"B33","doi-asserted-by":"publisher","first-page":"e104","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational radiomics system to decode the radiographic phenotype","volume":"77","author":"Van Griethuysen","year":"2017","journal-title":"Cancer Res"},{"key":"B34","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1186\/s13244-020-00887-2","article-title":"Radiomics in medical imaging\u2014\u201chow-to\u201d guide and critical reflection","volume":"11","author":"Van Timmeren","year":"2020","journal-title":"Insights Imaging"},{"key":"B35","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1007\/s41666-021-00113-8","article-title":"Conformal prediction in clinical medical sciences. J. Healthc. Inform","volume":"6","author":"Vazquez","year":"2022","journal-title":"Res"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2310.15161","article-title":"Sam-med3d: towards general-purpose segmentation models for volumetric medical images","author":"Wang","year":"2023","journal-title":"arXiv [preprint"},{"key":"B37","doi-asserted-by":"publisher","first-page":"2212","DOI":"10.1002\/mp.16132","article-title":"Towards reliable head and neck cancers locoregional recurrence prediction using delta-radiomics and learning with rejection option","volume":"50","author":"Wang","year":"2023","journal-title":"Med. Phys"},{"key":"B38","doi-asserted-by":"publisher","first-page":"5392","DOI":"10.1002\/mp.14388","article-title":"A multi-objective radiomics model for the prediction of locoregional recurrence in head and neck squamous cell cancer","volume":"47","author":"Wang","year":"2020","journal-title":"Med. Phys"},{"key":"B39","doi-asserted-by":"publisher","first-page":"955712","DOI":"10.3389\/fonc.2022.955712","article-title":"Predicting local persistence\/recurrence after radiation therapy for head and neck cancer from PET\/CT using a multi-objective, multi-classifier radiomics model","volume":"12","author":"Zhang","year":"2022","journal-title":"Front. Oncol"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1623393\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T05:26:29Z","timestamp":1757309189000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1623393\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,8]]},"references-count":39,"alternative-id":["10.3389\/frai.2025.1623393"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1623393","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,8]]},"article-number":"1623393"}}