{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T11:18:39Z","timestamp":1783423119664,"version":"3.54.6"},"reference-count":33,"publisher":"MDPI AG","issue":"13","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Impacted third molars may adversely affect adjacent second molars, leading to pathological conditions such as external root resorption and dental caries. Accurate assessment of these interactions is important for treatment planning and clinical decision-making. Although cone-beam computed tomography (CBCT) provides detailed three-dimensional imaging, image interpretation remains challenging. Recent advances in artificial intelligence have enabled automated radiographic analysis using deep learning methods. Methods: This retrospective study included 162 CBCT scans obtained from patients aged 18\u201375 years. A total of 306 third molar\u2013second molar units were evaluated. Based on radiographic findings, interactions were categorized as independent, contact, or resorption. Several deep learning architectures were developed and evaluated, including conventional convolutional neural networks (CNNs), attention-based CNNs, and Vision Transformer (ViT) models. Performance was assessed using standard classification metrics, and an ensemble approach was applied to improve predictive stability. Results: Attention-based and Transformer-based models generally outperformed conventional CNN architectures. These models achieved better discrimination among the defined classes and demonstrated superior overall performance. The ensemble model produced the most reliable results, achieving the highest macro-area under the curve (macro-AUC) values. Distinguishing contact cases from independent cases was the most challenging task, whereas resorption cases were identified more consistently across different models. Conclusions: Transformer-based deep learning models showed promising performance for CBCT-based assessment of third molar\u2013second molar interactions. Ensemble learning further improved classification reliability and robustness. These findings suggest that artificial intelligence-assisted systems may support early detection of third molar-related pathological changes and contribute to more accurate radiological evaluation and clinical decision-making.<\/jats:p>","DOI":"10.3390\/diagnostics16132123","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T09:53:46Z","timestamp":1783418026000},"page":"2123","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative Evaluation of Hybrid Attention-CNN and Vision Transformer Models for Multi-Class Classification of Third\u2013Second Molar Relationships on CBCT"],"prefix":"10.3390","volume":"16","author":[{"given":"Hazal","family":"Karsl\u0131o\u011flu","sequence":"first","affiliation":[{"name":"Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Mersin University, Mersin 33343, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8793-1486","authenticated-orcid":false,"given":"Jale","family":"Bekta\u015f","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering, Mersin University, Mersin 33343, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-8509-3523","authenticated-orcid":false,"given":"Lutfiye","family":"Sal","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Mersin University, Mersin 33343, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mert","family":"Durukan","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Faculty of Engineering, Mersin University, Mersin 33343, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5863-6990","authenticated-orcid":false,"given":"Mehmet Ozgur","family":"Ozemre","sequence":"additional","affiliation":[{"name":"Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Mersin University, Mersin 33343, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"22642","DOI":"10.1038\/s41598-024-73556-9","article-title":"Prevalence and pattern of third molars impaction in a large Yemeni sample: A retrospective study","volume":"14","author":"Alhajj","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_2","first-page":"58","article-title":"Prevalence of second molar external root resorption caused by mandibular third molars: A CBCT study","volume":"71","author":"Granja","year":"2023","journal-title":"Gen. Dent."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Siotou, K., Kouskouki, M.P., Christopoulou, I., Tsolakis, A.I., and Tsolakis, I.A. (2022). Frequency and local etiological factors of impaction of permanent teeth among 1400 patients in a Greek population. Dent. J., 10.","DOI":"10.3390\/dj10080150"},{"key":"ref_4","first-page":"e511","article-title":"The effect of impacted third molars on second molar external root resorption: A cross-sectional CBCT study","volume":"28","author":"Aydin","year":"2023","journal-title":"Med. Oral Patol. Oral Cir. Bucal"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"15447","DOI":"10.1038\/s41598-021-95024-4","article-title":"Machine learning to predict distal caries in mandibular second molars associated with impacted third molars","volume":"11","author":"Hur","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"23688","DOI":"10.1038\/s41598-025-93783-y","article-title":"Deep learning-based approach to third molar impaction analysis with clinical classifications","volume":"15","author":"Balel","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1016\/j.jds.2020.06.019","article-title":"Developments, application, and performance of artificial intelligence in dentistry: A systematic review","volume":"16","author":"Khanagar","year":"2021","journal-title":"J. Dent. Sci."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"105373","DOI":"10.1016\/j.jdent.2024.105373","article-title":"Deep learning method to automatically diagnose periodontal bone loss and periodontitis stage in dental panoramic radiograph","volume":"150","author":"Xue","year":"2024","journal-title":"J. Dent."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Celik, M.E. (2022). Deep learning based detection tool for impacted mandibular third molar teeth. Diagnostics, 12.","DOI":"10.3390\/diagnostics12040942"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Lo Casto, A., Spartivento, G., Benfante, V., Di Raimondo, R., Ali, M., Di Raimondo, D., Tuttolomondo, A., Stefano, A., Yezzi, A., and Comelli, A. (2023). Artificial intelligence for classifying the relationship between impacted third molar and mandibular canal on panoramic radiographs. Life, 13.","DOI":"10.3390\/life13071441"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Pirayesh, Z., Mohammad-Rahimi, H., Motamedian, S.R., Amini Afshar, S., Abbasi, R., Rohban, M.H., Mahdian, M., Ahsaie, M.G., and Alamdari, M.I. (2024). A hierarchical deep learning approach for diagnosing impacted canine-induced root resorption via cone-beam computed tomography. BMC Oral Health, 24.","DOI":"10.1186\/s12903-024-04718-4"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Zhang, X., Guo, E., Liu, X., Zhao, H., Yang, J., Li, W., Wu, W., and Sun, W. (2025). Enhancing furcation involvement classification on panoramic radiographs with vision transformers. BMC Oral Health, 25.","DOI":"10.1186\/s12903-025-05431-6"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s00521-025-11730-4","article-title":"An enhanced deep learning model for detection and classification of dental caries in panoramic radiographs","volume":"38","author":"Ozdemir","year":"2026","journal-title":"Neural Comput. Appl."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"11534","DOI":"10.1109\/TNNLS.2025.3550979","article-title":"TransXNet: Learning both global and local dynamics with a dual dynamic token mixer for visual recognition","volume":"36","author":"Lou","year":"2025","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1007\/s11424-022-2057-9","article-title":"Transformer-based deep learning network for tooth segmentation on panoramic radiographs","volume":"36","author":"Sheng","year":"2023","journal-title":"J. Syst. Sci. Complex."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Khurshid, Z., Faridoon, F., Silkosessak, O.C.U.D., Trachoo, V., Waqas, M., Hasan, S., and Porntaveetus, T. (2025). Multi-regional deep learning models for identifying dental restorations and prosthesis in panoramic radiographs. BMC Oral Health, 25.","DOI":"10.1186\/s12903-025-07138-0"},{"key":"ref_17","unstructured":"Winter, G.B. (1926). Impact of the Lower Third Molar on the Dental Arch, American Medical Book Company."},{"key":"ref_18","first-page":"330","article-title":"Impacted mandibular third molars: Classification and modified techniques for removal","volume":"39","author":"Pell","year":"1933","journal-title":"Dent. Dig."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1186\/s40537-019-0197-0","article-title":"A survey on image data augmentation for deep learning","volume":"6","author":"Shorten","year":"2019","journal-title":"J. Big Data"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Woo, S., Park, J., Lee, J.Y., and Kweon, I.S. (2018, January 8\u201314). CBAM: Convolutional block attention module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01234-2_1"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Hu, J., Shen, L., and Sun, G. (2018, January 18\u201323). Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1093\/dmfr\/twaf072","article-title":"Detection of external root resorption in periapical radiographs using YOLO-based deep learning model","volume":"55","author":"Gokdeniz","year":"2025","journal-title":"Dentomaxillofac. Radiol."},{"key":"ref_23","unstructured":"Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., and Gelly, S. (2020). An image is worth 16\u00d716 words: Transformers for image recognition at scale. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"964","DOI":"10.1109\/TNN.2007.897478","article-title":"Combining pattern classifiers: Methods and algorithms","volume":"18","author":"Alpaydin","year":"2007","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Dietterich, T.G. (2000). Ensemble methods in machine learning. Multiple Classifier Systems, Springer.","DOI":"10.1007\/3-540-45014-9_1"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"107241","DOI":"10.1016\/j.rineng.2025.107241","article-title":"Composite-metric deep learning for pathological complete response prediction from breast MRI","volume":"28","author":"Bahba","year":"2025","journal-title":"Results Eng."},{"key":"ref_27","unstructured":"Simonyan, K., Vedaldi, A., and Zisserman, A. (2013). Deep inside convolutional networks: Visualising image classification models and saliency maps. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22\u201329). Grad-CAM: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1016\/j.oooo.2023.02.021","article-title":"Developing deep learning methods for classification of teeth in dental panoramic radiography","volume":"138","author":"Yilmaz","year":"2024","journal-title":"Oral Surg. Oral Med. Oral Pathol. Oral Radiol."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"108755","DOI":"10.1016\/j.compbiomed.2024.108755","article-title":"AI-driven localization of all impacted teeth and prediction of Winter angulation for third molars on panoramic radiographs: Clinical user interface design","volume":"178","author":"Zirek","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"12609","DOI":"10.1038\/s41598-021-92121-2","article-title":"Classification of caries in third molars on panoramic radiographs using deep learning","volume":"11","author":"Vinayahalingam","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"109467","DOI":"10.1016\/j.identj.2026.109467","article-title":"A deep learning model for second-molar lesions related to impacted third molars","volume":"76","author":"Jiang","year":"2026","journal-title":"Int. Dent. J."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Adali, E., Ozden Yuce, M., Isik, G., Sener, E., and Mert, A. (2024). Treatment decision for impacted mandibular third molars: Effects of cone-beam computed tomography and level of surgeons\u2019 experience. PLoS ONE, 19.","DOI":"10.1371\/journal.pone.0314883"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/13\/2123\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T10:59:19Z","timestamp":1783421959000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/13\/2123"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,7]]},"references-count":33,"journal-issue":{"issue":"13","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["diagnostics16132123"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16132123","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,7]]}}}