{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,16]],"date-time":"2026-08-16T08:29:33Z","timestamp":1786868973992,"version":"build-2736575974"},"reference-count":31,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2025,9,27]],"date-time":"2025-09-27T00:00:00Z","timestamp":1758931200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"name":"Yonsei University College of Dentistry Fund","award":["6-2022-0023"],"award-info":[{"award-number":["6-2022-0023"]}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100014188","name":"MSIT","doi-asserted-by":"publisher","award":["IRIS RS-2024-00353147"],"award-info":[{"award-number":["IRIS RS-2024-00353147"]}],"id":[{"id":"10.13039\/501100014188","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Institute of Information & Communications Technology Planning & Evaluation","award":["RS-2022-00155966"],"award-info":[{"award-number":["RS-2022-00155966"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,2,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Inferior alveolar nerve (IAN) injury is a critical complication of mandibular third-molar extraction. This study aimed to construct and evaluate a deep learning framework that integrates contrastive learning and Bayesian optimization to enhance predictive performance using cone-beam computed tomography (CBCT) and panoramic radiographs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>A retrospective dataset of 902 panoramic radiographs and 1,500 CBCT images was used. Five deep learning architectures (MobileNetV2, ResNet101D, Vision Transformer, Twins-SVT, and SSL-ResNet50) were trained with and without contrastive learning and Bayesian optimization. Model performance was evaluated using accuracy, F1-score, and comparison with oral and maxillofacial surgeons (OMFSs).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Contrastive learning significantly improved the F1-scores across all models (e.g., MobileNetV2: 0.302 to 0.740; ResNet101D: 0.188 to 0.689; Vision Transformer: 0.275 to 0.704; Twins-SVT: 0.370 to 0.719; SSL-ResNet50: 0.109 to 0.576). Bayesian optimization further enhanced the F1-scores for MobileNetV2 (from 0.740 to 0.923), ResNet101D (from 0.689 to 0.857), Vision Transformer (from 0.704 to 0.871), Twins-SVT (from 0.719 to 0.857), and SSL-ResNet50 (from 0.576 to 0.875). The AI model outperformed OMFSs on CBCT cross-sectional images (F1-score: 0.923 vs. 0.667) but underperformed on panoramic radiographs (0.666 vs. 0.730).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed single-step deep learning approach effectively predicts IAN injury, with contrastive learning addressing data imbalance and Bayesian optimization enhancing model performance. While artificial intelligence surpasses human performance on CBCT images, panoramic radiograph analysis still benefits from expert interpretation. Future work should focus on multi-center validation and explainable artificial intelligence for broader clinical adoption.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/dmfr\/twaf069","type":"journal-article","created":{"date-parts":[[2025,9,27]],"date-time":"2025-09-27T07:40:31Z","timestamp":1758958831000},"page":"150-157","source":"Crossref","is-referenced-by-count":2,"title":["Single-step prediction of inferior alveolar nerve injury after mandibular third-molar extraction using contrastive learning and Bayesian auto-tuned deep learning model"],"prefix":"10.1093","volume":"55","author":[{"given":"Kyubaek","family":"Yoon","sequence":"first","affiliation":[{"name":"Department of Artificial Intelligence and Software, College of Artificial Intelligence, Ewha Womans University , Seoul 03766,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6283-9686","authenticated-orcid":false,"given":"Yiseul","family":"Choi","sequence":"additional","affiliation":[{"name":"Department of Advanced General Dentistry, Yonsei University College of Dentistry , Seoul 03722,","place":["Republic of Korea"]},{"name":"Yonsei Institute for Digital Health, Yonsei University , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Myoungho","family":"Lee","sequence":"additional","affiliation":[{"name":"Department of Advanced General Dentistry, Yonsei University College of Dentistry , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaeyeon","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Advanced General Dentistry, Yonsei University College of Dentistry , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun-Young","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Surgery, Yonsei University College of Dentistry , Seoul 03722,","place":["Republic of Korea"]},{"name":"Yonsei Institute for Digital Health, Yonsei University , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jin-Woo","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Surgery, School of Medicine, Ewha Womans University , Seoul 07804,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jongeun","family":"Choi","sequence":"additional","affiliation":[{"name":"School of Mechanical Engineering, Yonsei University , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2081-1156","authenticated-orcid":false,"given":"Wonse","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Advanced General Dentistry, Yonsei University College of Dentistry , Seoul 03722,","place":["Republic of Korea"]},{"name":"Yonsei Institute for Digital Health, Yonsei University , Seoul 03722,","place":["Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,9,27]]},"reference":[{"key":"2026020911294337400_twaf069-B1","doi-asserted-by":"publisher","first-page":"20190265","DOI":"10.1259\/dmfr.20190265","article-title":"Panoramic versus CBCT used to reduce inferior alveolar nerve paresthesia after third molar extractions: a systematic review and meta-analysis","volume":"49","author":"Del Lhano","year":"2020","journal-title":"Dentomaxillofac Radiol"},{"key":"2026020911294337400_twaf069-B2","doi-asserted-by":"publisher","first-page":"663","DOI":"10.1016\/j.joms.2016.12.013","article-title":"Predictive value of panoramic radiography for injury of inferior alveolar nerve after mandibular third molar surgery","volume":"75","author":"Su","year":"2017","journal-title":"J Oral Maxillofac Surg"},{"key":"2026020911294337400_twaf069-B3","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1016\/j.jormas.2019.06.010","article-title":"Determining the risk relationship associated with inferior alveolar nerve injury following removal of mandibular third molar teeth: a systematic review","volume":"121","author":"Kang","year":"2020","journal-title":"J Stomatol Oral Maxillofac Surg"},{"key":"2026020911294337400_twaf069-B4","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1007\/s00784-025-06285-6","article-title":"Automatic mandibular third molar and mandibular canal relationship determination based on deep learning models for preoperative risk reduction","volume":"29","author":"Yasin","year":"2025","journal-title":"Clin Oral Investig"},{"key":"2026020911294337400_twaf069-B5","first-page":"254","article-title":"Accuracy of panoramic radiographic predictor signs in the assessment of proximity of impacted third molars with the mandibular canal","volume":"13","author":"Elkhateeb","year":"2018","journal-title":"J Taibah Univ Med Sci"},{"key":"2026020911294337400_twaf069-B6","doi-asserted-by":"crossref","first-page":"181","DOI":"10.5624\/isd.20210289","article-title":"Inter-observer reliability in cone-beam computed tomography assessment of the retromolar canal: a practical plan to improve diagnostic imaging","volume":"52","author":"Igarashi","year":"2022","journal-title":"Imaging Sci Dent"},{"key":"2026020911294337400_twaf069-B7","doi-asserted-by":"crossref","first-page":"433","DOI":"10.1016\/j.cmpb.2013.10.015","article-title":"Teeth segmentation of dental periapical radiographs based on local singularity analysis","volume":"113","author":"Lin","year":"2014","journal-title":"Comput Methods Programs Biomed"},{"key":"2026020911294337400_twaf069-B8","doi-asserted-by":"crossref","first-page":"329","DOI":"10.1016\/j.jdsr.2023.09.004","article-title":"Deep learning and artificial intelligence in dental diagnostic imaging","volume":"59","author":"Katsumata","year":"2023","journal-title":"Jpn Dent Sci Rev"},{"key":"2026020911294337400_twaf069-B9","doi-asserted-by":"publisher","first-page":"2456","DOI":"10.1038\/s41598-022-06483-2","article-title":"Artificial intelligence in positioning between mandibular third molar and inferior alveolar nerve on panoramic radiography","volume":"12","author":"Choi","year":"2022","journal-title":"Sci Rep"},{"key":"2026020911294337400_twaf069-B10","doi-asserted-by":"publisher","first-page":"5842","DOI":"10.1038\/s41598-020-62321-3","article-title":"Deep learning method for mandibular canal segmentation in dental cone beam computed tomography volumes","volume":"10","author":"Jaskari","year":"2020","journal-title":"Sci Rep"},{"key":"2026020911294337400_twaf069-B11","doi-asserted-by":"publisher","first-page":"16925","DOI":"10.1038\/s41598-022-21408-9","article-title":"Deep learning model for analyzing the relationship between mandibular third molar and inferior alveolar nerve in panoramic radiography","volume":"12","author":"Sukegawa","year":"2022","journal-title":"Sci Rep"},{"key":"2026020911294337400_twaf069-B12","doi-asserted-by":"publisher","first-page":"981","DOI":"10.1007\/s00784-021-04082-5","article-title":"Deep learning-based evaluation of the relationship between mandibular third molar and mandibular canal on CBCT","volume":"26","author":"Liu","year":"2022","journal-title":"Clin Oral Investig"},{"key":"2026020911294337400_twaf069-B13","doi-asserted-by":"crossref","first-page":"292","DOI":"10.1016\/j.joms.2008.06.076","article-title":"Sensory testing of inferior alveolar nerve injuries: a review of methods used in prospective studies","volume":"67","author":"Poort","year":"2009","journal-title":"J Oral Maxillofac Surg"},{"key":"2026020911294337400_twaf069-B14","doi-asserted-by":"crossref","first-page":"125","DOI":"10.3390\/info11020125","article-title":"Albumentations: fast and flexible image augmentations","volume":"11","author":"Buslaev","year":"2020","journal-title":"Information"},{"key":"2026020911294337400_twaf069-B15","first-page":"3733","author":"Wu","year":"2018"},{"key":"2026020911294337400_twaf069-B16","volume-title":"Siamese Neural Networks for One-Shot Image Recognition","author":"Koch","year":"2015"},{"key":"2026020911294337400_twaf069-B17","first-page":"4510","author":"Sandler","year":"2018"},{"key":"2026020911294337400_twaf069-B18","first-page":"558","author":"He","year":"2019"},{"key":"2026020911294337400_twaf069-B19","author":"Houlsby ADLBAKDWXZTUMDMMGHSGJUN","year":"2021"},{"key":"2026020911294337400_twaf069-B20","doi-asserted-by":"publisher","first-page":"9355","DOI":"10.48550\/arXiv.2104.13840","article-title":"Twins: revisiting the design of spatial attention in vision transformers","volume":"34","author":"Chu","year":"2021","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026020911294337400_twaf069-B21","first-page":"22243","article-title":"Big self-supervised models are strong semi-supervised learners","volume":"33","author":"Chen","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026020911294337400_twaf069-B22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-016-0043-","article-title":"A survey of transfer learning","volume":"3","author":"Weiss","year":"2016","journal-title":"J Big Data"},{"key":"2026020911294337400_twaf069-B23","first-page":"248","volume-title":"Imagenet: A Large-Scale Hierarchical Image Database","author":"Deng","year":"2009"},{"key":"2026020911294337400_twaf069-B24","doi-asserted-by":"publisher","first-page":"25","DOI":"10.48550\/arXiv.1206.2944","article-title":"Practical Bayesian optimization of machine learning algorithms","author":"Snoek","year":"2012","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026020911294337400_twaf069-B25","first-page":"24:2546-2554","article-title":"Algorithms for hyper-parameter optimization","author":"Bergstra","year":"2011","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026020911294337400_twaf069-B26","first-page":"230","article-title":"A system for massively parallel hyperparameter tuning","volume":"2","author":"Li","year":"2020","journal-title":"Proc Mach Learn Syst"},{"key":"2026020911294337400_twaf069-B27","doi-asserted-by":"crossref","first-page":"1664","DOI":"10.3390\/diagnostics11091664","article-title":"Artificial intelligence model to detect real contact relationship between mandibular third molars and inferior alveolar nerve based on panoramic radiographs","volume":"11","author":"Zhu","year":"2021","journal-title":"Diagnostics"},{"key":"2026020911294337400_twaf069-B28","doi-asserted-by":"publisher","first-page":"475","DOI":"10.3390\/app12010475","article-title":"Automated prediction of extraction difficulty and inferior alveolar nerve injury for mandibular third molar using a deep neural network","volume":"12","author":"Lee","year":"2022","journal-title":"Appl Sci"},{"key":"2026020911294337400_twaf069-B29","doi-asserted-by":"publisher","first-page":"9007","DOI":"10.1038\/s41598-019-45487-3","article-title":"Automated detection of third molars and mandibular nerve by deep learning","volume":"9","author":"Vinayahalingam","year":"2019","journal-title":"Sci Rep"},{"key":"2026020911294337400_twaf069-B30","doi-asserted-by":"publisher","first-page":"991","DOI":"10.1016\/j.jds.2022.12.008","article-title":"Deep learning model for the automated evaluation of contact between the lower third molar and inferior alveolar nerve on panoramic radiography","volume":"18","author":"Takebe","year":"2023","journal-title":"J Dent Sci"},{"key":"2026020911294337400_twaf069-B31","first-page":"21137","author":"Cipriano","year":"2022"}],"container-title":["Dentomaxillofacial Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/dmfr\/advance-article-pdf\/doi\/10.1093\/dmfr\/twaf069\/64417398\/twaf069.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/dmfr\/article-pdf\/55\/2\/150\/64417398\/twaf069.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/dmfr\/article-pdf\/55\/2\/150\/64417398\/twaf069.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T16:35:32Z","timestamp":1770654932000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/dmfr\/article\/55\/2\/150\/8266699"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,27]]},"references-count":31,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,2,1]]}},"URL":"https:\/\/doi.org\/10.1093\/dmfr\/twaf069","relation":{},"ISSN":["0250-832X","1476-542X"],"issn-type":[{"value":"0250-832X","type":"print"},{"value":"1476-542X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,2]]},"published":{"date-parts":[[2025,9,27]]}}}