{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T23:59:44Z","timestamp":1788479984602,"version":"build-2803163510"},"reference-count":30,"publisher":"SAGE Publications","issue":"13","license":[{"start":{"date-parts":[[2023,11,9]],"date-time":"2023-11-09T00:00:00Z","timestamp":1699488000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["J Dent Res"],"published-print":{"date-parts":[[2023,12]]},"abstract":"<jats:p>Machine learning (ML) models, especially deep neural networks, are increasingly being used for the analysis of medical images and as a supporting tool for clinical decision-making. In this study, we propose an artificial intelligence system to facilitate dental decision-making for the removal of mandibular third molars (M3M) based on 2-dimensional orthopantograms and the risk assessment of such a procedure. A total of 4,516 panoramic radiographic images collected at the Center of Dental Medicine at the University of Zurich, Switzerland, were used for training the ML model. After image preparation and preprocessing, a spatially dependent U-Net was employed to detect and retrieve the region of the M3M and inferior alveolar nerve (IAN). Image patches identified to contain a M3M were automatically processed by a deep neural network for the classification of M3M superimposition over the IAN (task 1) and M3M root development (task 2). A control evaluation set of 120 images, collected from a different data source than the training data and labeled by 5 dental practitioners, was leveraged to reliably evaluate model performance. By 10-fold cross-validation, we achieved accuracy values of 0.94 and 0.93 for the M3M\u2013IAN superimposition task and the M3M root development task, respectively, and accuracies of 0.9 and 0.87 when evaluated on the control data set, using a ResNet-101 trained in a semisupervised fashion. Matthew\u2019s correlation coefficient values of 0.82 and 0.75 for task 1 and task 2, evaluated on the control data set, indicate robust generalization of our model. Depending on the different label combinations of task 1 and task 2, we propose a diagnostic table that suggests whether additional imaging via 3-dimensional cone beam tomography is advisable. Ultimately, computer-aided decision-making tools benefit clinical practice by enabling efficient and risk-reduced decision-making and by supporting less experienced practitioners before the surgical removal of the M3M.<\/jats:p>","DOI":"10.1177\/00220345231200786","type":"journal-article","created":{"date-parts":[[2023,11,9]],"date-time":"2023-11-09T18:57:30Z","timestamp":1699556250000},"page":"1452-1459","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":22,"title":["Preinterventional Third-Molar Assessment Using Robust Machine Learning"],"prefix":"10.1177","volume":"102","author":[{"given":"J.S.","family":"Carvalho","sequence":"first","affiliation":[{"name":"ETH Zurich, Department of Computer Science, Zurich, Switzerland"},{"name":"ETH AI Center, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"M.","family":"Lotz","sequence":"additional","affiliation":[{"name":"University of Zurich, Center for Dental Medicine, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"L.","family":"Rubi","sequence":"additional","affiliation":[{"name":"ETH Zurich, Department of Computer Science, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S.","family":"Unger","sequence":"additional","affiliation":[{"name":"University of Zurich, Center for Dental Medicine, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"T.","family":"Pfister","sequence":"additional","affiliation":[{"name":"University of Zurich, Center for Dental Medicine, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J.M.","family":"Buhmann","sequence":"additional","affiliation":[{"name":"ETH Zurich, Department of Computer Science, Zurich, Switzerland"},{"name":"ETH AI Center, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5044-7052","authenticated-orcid":false,"given":"B.","family":"Stadlinger","sequence":"additional","affiliation":[{"name":"University of Zurich, Center for Dental Medicine, Zurich, Switzerland"},{"name":"ETH AI Center, Zurich, Switzerland"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2023,11,9]]},"reference":[{"key":"bibr1-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0177678"},{"key":"bibr2-00220345231200786","unstructured":"Carvalho JBS, Santinha J, Miladinovi\u0107 \u00d0, Cotrini C, Buhmann JM. 2022. Holistic modeling in medical image segmentation using spatial recurrence. Paper presented at: MIDL 2022. Proceedings of the 5th International Conference on Medical Imaging with Deep Learning. PLMR. 172:199\u2013218; Z\u00fcrich, Switzerland. [accessed 2023 July 12]; https:\/\/proceedings.mlr.press\/v172\/carvalho22a.html."},{"key":"bibr3-00220345231200786","unstructured":"Chen T, Kornblith S, Norouzi M, Hinton G. 2020. A simple framework for contrastive learning of visual representations. Paper presented at: PMLR 2020. Proceedings of the 37th International Conference on Machine Learning; Vienna, Austria. [accessed 2023 July 12]; http:\/\/proceedings.mlr.press\/v119\/chen20j\/chen20j.pdf."},{"key":"bibr4-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1186\/s12864-019-6413-7"},{"key":"bibr5-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-06483-2"},{"issue":"9","key":"bibr6-00220345231200786","doi-asserted-by":"crossref","first-page":"945","DOI":"10.61872\/sdj-2015-09-01","volume":"125","author":"Dula K","year":"2015","journal-title":"Swiss Dent J"},{"key":"bibr7-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1016\/j.joms.2010.10.039"},{"key":"bibr8-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1016\/j.oooo.2020.04.005"},{"key":"bibr9-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"bibr10-00220345231200786","unstructured":"Hendrycks D, Mazeika M, Kadavath S, Song D. 2019. Using self-supervised learning can improve model robustness and uncertainty. Adv Neural Inf Process Sys. 32. 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada. [accessed 2023 July 12]; https:\/\/arxiv.org\/pdf\/1906.12340.pdf."},{"key":"bibr11-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-09954-8"},{"key":"bibr12-00220345231200786","unstructured":"Jocher G. 2020. Yolov5 by ultralytics. In: software C, editor. 7.0 ed. [accessed 2022 Sep 7]. https:\/\/github.com\/ultralytics\/yolov5."},{"key":"bibr13-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1177\/00220345211032524"},{"key":"bibr14-00220345231200786","doi-asserted-by":"publisher","DOI":"10.3390\/app12010475"},{"key":"bibr15-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijom.2010.09.005"},{"key":"bibr16-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2017.07.005"},{"key":"bibr17-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1007\/s00784-021-04082-5"},{"key":"bibr18-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1177\/00220345221106086"},{"key":"bibr19-00220345231200786","unstructured":"Navarro F, Watanabe C, Shit S, Sekuboyina A, Peeken J, Combs SE, Menze BH. 2021. Evaluating the robustness of self-supervised learning in medical imaging. arXiv preprint arXiv:2105.06986. [accessed 2023 July 12]; https:\/\/arxiv.org\/abs\/2105.06986."},{"key":"bibr20-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1109\/tmm.2021.3109419"},{"key":"bibr21-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"bibr22-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1177\/0022034520915714"},{"key":"bibr23-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1177\/0022034520969115"},{"key":"bibr24-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2022.105458"},{"key":"bibr25-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-bioeng-071516-044442"},{"issue":"10","key":"bibr26-00220345231200786","first-page":"1042","volume":"124","author":"Sigron GR","year":"2014","journal-title":"Swiss Dent J"},{"key":"bibr27-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-21408-9"},{"key":"bibr28-00220345231200786","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez A, Kaiser \u0141. 2017. Attention is all you need. Adv Neural Inf Process Syst. 30. [accessed 2023 July 12]; https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2017\/file\/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf."},{"key":"bibr29-00220345231200786","doi-asserted-by":"publisher","DOI":"10.1177\/00220345211005338"},{"key":"bibr30-00220345231200786","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics11091664"}],"container-title":["Journal of Dental Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/00220345231200786","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/00220345231200786","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/00220345231200786","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T21:08:32Z","timestamp":1777496912000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/00220345231200786"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,9]]},"references-count":30,"journal-issue":{"issue":"13","published-print":{"date-parts":[[2023,12]]}},"alternative-id":["10.1177\/00220345231200786"],"URL":"https:\/\/doi.org\/10.1177\/00220345231200786","relation":{},"ISSN":["0022-0345","1544-0591"],"issn-type":[{"value":"0022-0345","type":"print"},{"value":"1544-0591","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,9]]}}}