{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T11:29:27Z","timestamp":1787138967915,"version":"build-2736575974"},"reference-count":46,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Scientific Research Projects Coordination Unit of Yozgat Bozok University","award":["1728-THD-2025-1728"],"award-info":[{"award-number":["1728-THD-2025-1728"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Third molar development is one of the most frequently used dental indicators in forensic age assessment because its maturation continues through adolescence and early adulthood. Manual staging on panoramic radiographs requires experience and may be affected by observer variability, especially in transitional developmental stages. This study evaluated the performance of segmentation-based deep learning models for automatic third molar localization and Demirjian-based developmental stage classification on panoramic radiographs. Methods: The study used a two-stage segmentation framework. In the first stage, third molar localization was performed using 737 panoramic radiographs and 2736 annotations for teeth 18, 28, 38, and 48. In the second stage, developmental stage classification was performed using 695 panoramic radiographs and 2533 annotations grouped as AB, CD, EF, and GH according to Demirjian developmental stages. A supplementary 18-year threshold segmentation analysis was added using 695 panoramic radiographs and 2573 labels divided into training, validation, and test sets at an 80:10:10 ratio. YOLO-based segmentation models were trained using AdamW optimization and learning-rate settings recorded in the training logs. Model performance was evaluated using TP, FP, FN, precision, recall, F1-score, accuracy, mAP, Dice coefficient, Jaccard index, and performance curves. For instance segmentation, true negatives were not calculated because the number of background non-objects is not finite or clinically meaningful. Results: The third molar localization model yielded TP = 256, FP = 4, and FN = 11 on the test set. The overall accuracy was 0.9446, mAP@0.5 was 0.980, mAP@0.5:0.95 was 0.875, Dice coefficient was 0.9064, and Jaccard index was 0.8696. In developmental stage classification, yolo11x-seg produced the most balanced segmentation profile, with an accuracy of 0.7272, Dice coefficient of 0.5783, and Jaccard index of 0.5529. In the supplementary 18-year threshold segmentation analysis, yolov8x-seg produced TP = 172, FP = 37, and FN = 32. Overall accuracy was 0.8152, precision was 0.8230, recall was 0.8431, F1-score was 0.8329, mAP@0.5 was 0.635, and mAP@0.5:0.95 was 0.563. Class-wise results showed better performance for the under-18 class than for the 18-years-and-older class. Conclusions: YOLO-based segmentation models can localize third molars on panoramic radiographs with high performance. Developmental stage classification is more difficult than anatomical localization because the radiographic boundaries between adjacent developmental stages are gradual rather than discrete. The added 18-year threshold analysis provides a clinically relevant age threshold experiment, but it also shows that class imbalance and weak segmentation of the 18-years-and-older class limit direct forensic use. Segmentation-based third molar analysis should therefore be interpreted as a visually auditable decision-support workflow rather than a stand-alone legal age determination tool.<\/jats:p>","DOI":"10.3390\/diagnostics16162636","type":"journal-article","created":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T11:04:12Z","timestamp":1787137452000},"page":"2636","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative Analysis of Third Molar Segmentation Performance Across Dental Developmental Stages and the 18-Year Age Threshold Using Deep Learning Models"],"prefix":"10.3390","volume":"16","author":[{"given":"Melis B\u00fc\u015fra","family":"A\u015fk\u0131n","sequence":"first","affiliation":[{"name":"Department of Orthodontics, Faculty of Dentistry, Yozgat Bozok University, 66100 Yozgat, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ay\u015fe","family":"Bulut","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Radiology, Physiology, Faculty of Dentistry, Yozgat Bozok University, 66100 Yozgat, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0818-6746","authenticated-orcid":false,"given":"G\u00f6kalp","family":"\u00c7\u0131narer","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering-Architecture, Yozgat Bozok University, 66100 Yozgat, Turkey"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,19]]},"reference":[{"key":"ref_1","first-page":"211","article-title":"A new system of dental age assessment","volume":"45","author":"Demirjian","year":"1973","journal-title":"Hum. Biol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1080\/03014467600001671","article-title":"New systems for dental maturity based on seven and four teeth","volume":"3","author":"Demirjian","year":"1976","journal-title":"Ann. Hum. Biol."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Neves, J.A., Lopes, B.D., Machado, V., Delgado, A.S., and Mendes, J.J. (2024). Evidence of Age Estimation Procedures in Forensic Dentistry: Results from an Umbrella Review. Medicina, 60.","DOI":"10.3390\/medicina60010042"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"8783660","DOI":"10.1155\/2024\/8783660","article-title":"Radiological evaluation of the accuracy of Demirjian, Nolla, and Willems methods for dental age estimation in 3\u201317-year-old Iranian children","volume":"2024","author":"Pliska","year":"2024","journal-title":"Biomed. Res. Int."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Stamm, A.D.B., Cariego, M.T., Vazquez, D.J., Pujol, M.H., Saiegh, J., Bielli, M.V., Hetch, P., Carosi, M.J., and Cabirta, M.L. (2022). Use of the Demirjian method to estimate dental age in panoramic radiographs of patients treated at the Buenos Aires University School of Dentistry. Acta Odontol. Latinoam., 35.","DOI":"10.54589\/aol.35\/1\/25"},{"key":"ref_6","first-page":"28","article-title":"Age estimation at 18-year threshold: Comparing Demirjian and Cameriere\u2019s methods for Thais","volume":"42","author":"Upalananda","year":"2024","journal-title":"J. Forensic Odontostomatol."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Caggiano, M., Scelza, G., Amato, A., Orefice, R., Belli, S., Pagano, S., Valenti, C., and Martina, S. (2022). Estimating the 18-Year Threshold with Third Molars Radiographs in the Southern Italy Population: Accuracy and Reproducibility of Demirjian Method. Int. J. Environ. Res. Public Health, 19.","DOI":"10.3390\/ijerph191610454"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2481","DOI":"10.1007\/s00414-024-03295-z","article-title":"Chronological age estimation based on dental mineralization for Syrian population","volume":"138","author":"Alissa","year":"2024","journal-title":"Int. J. Leg. Med."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"153","DOI":"10.4103\/0975-1475.155081","article-title":"Demirjian\u2019s method in the estimation of age: A study on human third molars","volume":"7","author":"Lewis","year":"2015","journal-title":"J. Forensic Dent. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"177","DOI":"10.4103\/0975-1475.195103","article-title":"Demirjian approach of dental age estimation: Abridged for operator ease","volume":"8","author":"Jain","year":"2016","journal-title":"J. Forensic Dent. Sci."},{"key":"ref_11","first-page":"e160","article-title":"Is chronological age predictable on panoramic images using the Demirjian method on third molars? Insights from a Turkish population-based study","volume":"31","year":"2026","journal-title":"Med. Oral Patol. Oral Cir. Bucal"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"2411","DOI":"10.1007\/s00414-024-03278-0","article-title":"Accuracy of second and third molar maturity indices, Olze, Demirjian and Haavikko methods for age estimation","volume":"138","author":"Shi","year":"2024","journal-title":"Int. J. Leg. Med."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1007\/s12024-025-01042-3","article-title":"Dental age estimation by comparing Demirjian\u2019s method and machine learning in Southeast Brazilian youth","volume":"22","author":"Abuabara","year":"2026","journal-title":"Forensic Sci. Med. Pathol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"4668","DOI":"10.1038\/s41598-024-55497-5","article-title":"Binary decisions of artificial intelligence to classify third molar development around the legal age thresholds of 14, 16 and 18 years","volume":"14","author":"Franco","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"13082","DOI":"10.1038\/s41598-024-63744-y","article-title":"A fully automated classification of third molar development stages using deep learning","volume":"14","author":"Milani","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"2469","DOI":"10.1007\/s00414-024-03298-w","article-title":"Artificial intelligence and dental age estimation: Development and validation of an automated stage allocation technique on all mandibular tooth types in panoramic radiographs","volume":"138","author":"Matthijs","year":"2024","journal-title":"Int. J. Leg. Med."},{"key":"ref_17","first-page":"143","article-title":"Automatic dental age calculation from panoramic radiographs using deep learning techniques","volume":"24","author":"Kokomoto","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Balel, Y., Sa\u011fta\u015f, K., and B\u00fclb\u00fcl, H.N. (2025). Development and evaluation of a deep learning-based system for dental age estimation using the Demirjian method on panoramic radiographs. BMC Oral Health, 25.","DOI":"10.1186\/s12903-025-06420-5"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Salehizeinabadi, M., Shrout, M., Hassanzadeh, A., and Amin, M. (2025). Dental age prediction from panoramic radiographs using machine learning techniques. PLoS Digit. Health, 4.","DOI":"10.1371\/journal.pdig.0001077"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"105560","DOI":"10.1016\/j.jdent.2025.105560","article-title":"Deep learning for age estimation from panoramic radiographs: A systematic review and meta-analysis","volume":"154","author":"Rokhshad","year":"2025","journal-title":"J. Dent."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Khanagar, S.B., Albalawi, F., Alshehri, A., Awawdeh, M., Iyer, K., Alsomaie, B., Aldhebaib, A., Singh, O.G., and Alfadley, A. (2024). Performance of artificial intelligence models designed for automated estimation of age using dento-maxillofacial radiographs: A systematic review. Diagnostics, 14.","DOI":"10.3390\/diagnostics14111079"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"101010","DOI":"10.1016\/j.ortho.2025.101010","article-title":"Efficacy of artificial intelligence in radiographic dental age estimation of patients undergoing dental maturation: A systematic review and meta-analysis","volume":"23","author":"Shahbazi","year":"2025","journal-title":"Int. Orthod."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1534","DOI":"10.1016\/j.jobcr.2025.09.010","article-title":"Artificial intelligence in dental age estimation: Applications, technological advances and legal aspects: A narrative review","volume":"15","author":"Chopra","year":"2025","journal-title":"J. Oral Biol. Craniofacial Res."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Bonfanti-Gris, M., Herrera, A., Salido, M.P., Mart\u00ednez-Rus, F., and Prad\u00edes, G. (2025). Deep learning for tooth detection and segmentation in panoramic radiographs: A systematic review and meta-analysis. BMC Oral Health, 25.","DOI":"10.1186\/s12903-025-06349-9"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Incerti Parenti, S., Tsiotas, G., Maglioni, A., Lamberti, G., Fiordelli, A., Rossi, D., Bononi, L., and Alessandri-Bonetti, G. (2025). Artificial intelligence-aided tooth detection and segmentation on pediatric panoramic radiographs in mixed dentition using a transfer learning approach. Diagnostics, 15.","DOI":"10.3390\/diagnostics15202615"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"1461","DOI":"10.29271\/jcpsp.2025.11.1461","article-title":"A deep learning approach for detecting periapical lesions on panoramic radiographic images","volume":"35","author":"Pekiner","year":"2025","journal-title":"J. Coll. Physicians Surg. Pak."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Ong, S.H., Kim, H., Song, J.S., Shin, T.J., Hyun, H.K., Jang, K.T., and Kim, Y.J. (2024). Fully automated deep learning approach to dental development assessment in panoramic radiographs. BMC Oral Health, 24.","DOI":"10.1186\/s12903-024-04160-6"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Peker, R.B., and Kurtoglu, C.O. (2025). Evaluation of the performance of a YOLOv10-based deep learning model for tooth detection and numbering on panoramic radiographs of patients in the mixed dentition period. Diagnostics, 15.","DOI":"10.3390\/diagnostics15040405"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Beser, B., Reis, T., Berber, M.N., Topaloglu, E., Gungor, E., K\u0131l\u0131c, M.C., Duman, S., \u00c7elik, \u00d6., Kuran, A., and Bayrakdar, I.S. (2024). YOLO-V5 based deep learning approach for tooth detection and segmentation on pediatric panoramic radiographs in mixed dentition. BMC Med. Imaging, 24.","DOI":"10.1186\/s12880-024-01410-5"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"537","DOI":"10.1016\/j.oooo.2023.06.003","article-title":"Automated permanent tooth detection and numbering on panoramic radiograph using a deep learning approach","volume":"137","author":"Putra","year":"2024","journal-title":"Oral Surg. Oral Med. Oral Pathol. Oral Radiol."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"2257","DOI":"10.1007\/s00784-020-03544-6","article-title":"Artificial intelligence-driven novel tool for tooth detection and segmentation on panoramic radiographs","volume":"25","author":"Leite","year":"2021","journal-title":"Clin. Oral Investig."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Bilgir, E., Bayrakdar, \u0130.\u015e., \u00c7elik, \u00d6., Orhan, K., Akkoca, F., Sa\u011flam, H., Odaba\u015f, A., Aslan, A.F., Ozcetin, C., and K\u0131ll\u0131, M. (2021). An artificial intelligence approach to automatic tooth detection and numbering in panoramic radiographs. BMC Med. Imaging, 21.","DOI":"10.1186\/s12880-021-00656-7"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"883","DOI":"10.33808\/clinexphealthsci.1219160","article-title":"A deep learning approach to automatic tooth detection and numbering in panoramic radiographs: An artificial intelligence study","volume":"13","author":"Keser","year":"2023","journal-title":"Clin. Exp. Health Sci."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Akdo\u011fan, S., \u00d6zi\u00e7, M.\u00dc., and Tassoker, M. (2025). Development of an AI-supported clinical tool for assessing mandibular third molar tooth extraction difficulty using panoramic radiographs and YOLO11 sub-models. Diagnostics, 15.","DOI":"10.3390\/diagnostics15040462"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"452","DOI":"10.1016\/j.ijom.2024.11.008","article-title":"Development and validation of a deep learning algorithm for the classification of the level of surgical difficulty in impacted mandibular third molar surgery","volume":"54","author":"Chindanuruks","year":"2025","journal-title":"Int. J. Oral Maxillofac. Surg."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1954","DOI":"10.1038\/s41598-021-81449-4","article-title":"Deep learning based prediction of extraction difficulty for mandibular third molars","volume":"11","author":"Yoo","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"38814","DOI":"10.1038\/s41598-025-22707-7","article-title":"End-to-end CNN-based detection of permanent first molars and prediction of root development stages from panoramic radiographs","volume":"15","author":"Kayaci","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"635","DOI":"10.1016\/j.oooo.2019.11.007","article-title":"Application of a fully deep convolutional neural network to the automation of tooth segmentation on panoramic radiographs","volume":"129","author":"Lee","year":"2020","journal-title":"Oral Surg. Oral Med. Oral Pathol. Oral Radiol."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Tuzoff, D.V., Tuzova, L.N., Bornstein, M.M., Krasnov, A.S., Kharchenko, M.A., Nikolenko, S.I., Sveshnikov, M.M., and Bednenko, G.B. (2019). Tooth detection and numbering in panoramic radiographs using convolutional neural networks. Dentomaxillofacial Radiol., 48.","DOI":"10.1259\/dmfr.20180051"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Chaipattanawan, N., Khwankaew, K., Chormanee, K., Chinkanjanaroj, A., and Pethpand, T. (2026). Demirjian\u2019s and Cameriere\u2019s methods for estimating the 18-year legal age threshold using mandibular third molars. BMC Oral Health, 16.","DOI":"10.1038\/s41598-026-43185-5"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Saito, T., and Rehmsmeier, M. (2015). The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0118432"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"103819","DOI":"10.1016\/j.media.2025.103819","article-title":"Knowledge distillation and teacher-student learning in medical imaging: Comprehensive overview, pivotal role, and future directions","volume":"107","author":"Li","year":"2026","journal-title":"Med. Image Anal."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"8712","DOI":"10.1109\/JIOT.2025.3641888","article-title":"PHH-FL: Perceptual Hashing Hypernetwork Personalized Federated Learning for Heterogeneous Medical Image Analysis Tasks","volume":"13","author":"Sun","year":"2026","journal-title":"IEEE Internet Things J."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"33110","DOI":"10.1109\/JIOT.2026.3694933","article-title":"Optical Fiber Intelligent Carpet for Gait Recognition with A Local Strain and Global Contour Dual-modality Attention Network","volume":"13","author":"Li","year":"2026","journal-title":"IEEE Internet Things J."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"e240300","DOI":"10.1148\/ryai.240300","article-title":"Checklist for Artificial Intelligence in Medical Imaging: CLAIM 2024 update","volume":"6","author":"Tejani","year":"2024","journal-title":"Radiol. Artif. Intell."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/16\/2636\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T11:19:16Z","timestamp":1787138356000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/16\/2636"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,19]]},"references-count":46,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["diagnostics16162636"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16162636","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,19]]}}}