{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T10:02:38Z","timestamp":1785319358605,"version":"3.55.0"},"reference-count":30,"publisher":"Oxford University Press (OUP)","issue":"4","license":[{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>To develop and validate an automated pipeline for maxillary dental midline localization on posed smile photographs and compare the performance of four convolutional neural network (CNN) backbones.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>A total of 221 standardized frontal posed smile photographs were collected. The maxillary dental midline was annotated by two orthodontists across two sessions, and mean coordinates served as the orthodontist-derived reference standard. An automated pipeline including face detection, interpupillary-line rotational normalization, mouth region extraction, and CNN-based keypoint regression was developed. Four CNN models (EfficientNet-B0, ResNet-18, MobileNetV3-Small, and HRNet-W18) were trained and compared using 5-fold stratified cross-validation. Deviation was expressed as the angle between the predicted dental midline and the interpupillary-derived photographic reference line (T-line).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>HRNet-W18 achieved the lowest mean angular error (0.63\u00b0 \u00b1 0.63\u00b0) and the highest \u22641\u00b0 error-threshold accuracy (81.0%). All 4 models exceeded 90% accuracy at the \u22642\u00b0 threshold. No significant effect of tooth visibility grade on angular error was observed. HRNet-W18 also showed the highest agreement with the reference standard [intraclass correlation coefficient (ICC) = 0.721] and minimal Bland\u2013Altman bias (+0.03\u00b0). Post hoc pairwise comparisons showed that HRNet-W18 outperformed all other models across all error metrics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Limitations<\/jats:title>\n                    <jats:p>The sample was limited to standardized posed smile photographs from a single center, and generalizability to diverse populations and less controlled conditions remains to be established.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions and Implications<\/jats:title>\n                    <jats:p>The automated pipeline showed high accuracy for dental midline localization on posed smile photographs relative to an orthodontist-derived reference standard. Further validation on larger and more diverse datasets is needed before clinical application.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/ejo\/cjag056","type":"journal-article","created":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T11:54:56Z","timestamp":1783684496000},"source":"Crossref","is-referenced-by-count":0,"title":["Dental midline localization: development and validation of an automated deep learning pipeline"],"prefix":"10.1093","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-8920-8605","authenticated-orcid":false,"given":"Arda","family":"Ar\u0131san","sequence":"first","affiliation":[{"name":"Independent Researcher , Ankara 06510 ,","place":["Turkey"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6152-6178","authenticated-orcid":false,"given":"G\u00f6khan Serhat","family":"Duran","sequence":"additional","affiliation":[{"name":"Department of Orthodontics, Faculty of Dentistry, \u00c7anakkale Onsekiz Mart University , \u00c7anakkale 17100 ,","place":["Turkey"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"2026072905282772900_cjag056-B1","doi-asserted-by":"publisher","first-page":"405","DOI":"10.2174\/1874210602115010405","article-title":"Dental-facial midline: an esthetic based classification","volume":"15","author":"Niraula","year":"2021","journal-title":"Open Dent J"},{"key":"2026072905282772900_cjag056-B2","doi-asserted-by":"publisher","first-page":"311","DOI":"10.1111\/j.1708-8240.1999.tb00414.x","article-title":"Comparing the perception of dentists and lay people to altered dental esthetics","volume":"11","author":"Kokich","year":"1999","journal-title":"J Esthet Dent"},{"key":"2026072905282772900_cjag056-B3","doi-asserted-by":"publisher","first-page":"e1053","DOI":"10.4317\/jced.61930","article-title":"Coincidence of the facial and maxillary dental midlines in the Peruvian population. A cross-sectional study","volume":"16","author":"Chahuara-Ram\u00edrez","year":"2024","journal-title":"J Clin Exp Dent"},{"key":"2026072905282772900_cjag056-B4","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1111\/ocr.12539","article-title":"Evaluation of an automated approach for facial midline detection and asymmetry assessment: a preliminary study","volume":"24 Suppl 2","author":"Yurdakurban","year":"2021","journal-title":"Orthod Craniofac Res"},{"key":"2026072905282772900_cjag056-B5","doi-asserted-by":"publisher","first-page":"515","DOI":"10.2319\/052209-286.1","article-title":"Young people\u2019s esthetic perception of dental midline deviation","volume":"80","author":"Zhang","year":"2010","journal-title":"Angle Orthod"},{"key":"2026072905282772900_cjag056-B6","doi-asserted-by":"publisher","DOI":"10.3390\/jcm13133944","article-title":"Impact of dental midline shift on the perception of facial attractiveness in young adults","volume":"13","author":"Sayahpour","year":"2024","journal-title":"J Clin Med"},{"key":"2026072905282772900_cjag056-B7","doi-asserted-by":"publisher","first-page":"748","DOI":"10.1016\/j.ajodo.2006.01.039","article-title":"Impact of dental asymmetries on the perception of smile esthetics","volume":"132","author":"Pinho","year":"2007","journal-title":"Am J Orthod Dentofacial Orthop"},{"key":"2026072905282772900_cjag056-B8","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1093\/ejo\/21.5.517","article-title":"The influence of dental to facial midline discrepancies on dental attractiveness ratings","volume":"21","author":"Johnston","year":"1999","journal-title":"Eur J Orthod"},{"key":"2026072905282772900_cjag056-B9","doi-asserted-by":"publisher","DOI":"10.3390\/jaestheticmed1010004","article-title":"AI-driven aesthetic rehabilitation in edentulous arches: advancing symmetry and smile design through Medit SmartX and Scan Ladder","volume":"1","author":"Nulty","year":"2025","journal-title":"J Aesthet Med"},{"key":"2026072905282772900_cjag056-B10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.31579\/2690-1919\/372","article-title":"Using artificial intelligence-based 3D smile design: REBEL in prosthetic rehabilitation with patient\u2019s personal preferences. Case-2","volume":"15","author":"Al-Maghreby","year":"2024","journal-title":"J Clin Res Rep"},{"key":"2026072905282772900_cjag056-B11","doi-asserted-by":"publisher","first-page":"312","DOI":"10.1016\/j.ortho.2019.03.014","article-title":"Quantification of maxillary dental midline deviation in 2D photographs: methodology trial","volume":"17","author":"Alarabi","year":"2019","journal-title":"Int Orthod"},{"key":"2026072905282772900_cjag056-B12","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1016\/S0022-3913(09)60117-7","article-title":"The relationship of facial anatomic landmarks with midlines of the face and mouth","volume":"102","author":"Bidra","year":"2009","journal-title":"J Prosthet Dent"},{"key":"2026072905282772900_cjag056-B13","doi-asserted-by":"publisher","first-page":"200","DOI":"10.5152\/TurkJOrthod.2019.18086","article-title":"Assessment of the relationship between facial and dental midlines with anatomical landmarks of the face and oral cavity","volume":"32","author":"Farahani","year":"2019","journal-title":"Turk J Orthod"},{"key":"2026072905282772900_cjag056-B14","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1007\/s00784-024-06021-6","article-title":"Accuracy of web-based automated versus digital manual cephalometric landmark identification","volume":"28","author":"Sadek","year":"2024","journal-title":"Clin Oral Investig"},{"key":"2026072905282772900_cjag056-B15","doi-asserted-by":"publisher","first-page":"362","DOI":"10.2319\/122324-1049.1","article-title":"Accuracy and reliability of automated landmark identification and cephalometric measurements on cone beam computed tomography using Invivo software","volume":"95","author":"Jung","year":"2025","journal-title":"Angle Orthod"},{"key":"2026072905282772900_cjag056-B16","doi-asserted-by":"publisher","first-page":"4005","DOI":"10.3390\/jcm9124005","article-title":"Digital dentistry: advances and challenges","volume":"9","author":"Schwendicke","year":"2020","journal-title":"J Clin Med"},{"key":"2026072905282772900_cjag056-B17","doi-asserted-by":"publisher","DOI":"10.3390\/jcm11030852","article-title":"Evaluating the precision of automatic segmentation of teeth, gingiva and facial landmarks for 2D digital smile design using real-time instance segmentation network","volume":"11","author":"Lee","year":"2022","journal-title":"J Clin Med"},{"key":"2026072905282772900_cjag056-B18","doi-asserted-by":"publisher","DOI":"10.3389\/frvir.2025.1652074","article-title":"A critical appraisal of computer vision in orthodontics","volume":"6","author":"Amm","year":"2025","journal-title":"Front Virtual Real"},{"key":"2026072905282772900_cjag056-B19","doi-asserted-by":"publisher","first-page":"110","DOI":"10.1016\/j.ajodo.2025.09.009","article-title":"Clinical advances in curve of Spee assessment: deep learning for automatic tooth landmark detection in Invisalign","volume":"169","author":"Ma","year":"2026","journal-title":"Am J Orthod Dentofacial Orthop"},{"key":"2026072905282772900_cjag056-B20","doi-asserted-by":"publisher","first-page":"1414","DOI":"10.3390\/sym14071414","article-title":"A review of three-dimensional facial asymmetry analysis methods","volume":"14","author":"Zhu","year":"2022","journal-title":"Symmetry (Basel)"},{"key":"2026072905282772900_cjag056-B21","doi-asserted-by":"publisher","first-page":"4705","DOI":"10.1007\/s00266-025-04857-1","article-title":"Evaluation of facial asymmetry determined by linear measurements according to hand preference and gender","volume":"49","author":"Barut","year":"2025","journal-title":"Aesthetic Plast Surg"},{"key":"2026072905282772900_cjag056-B22","author":"Feng"},{"key":"2026072905282772900_cjag056-B23","doi-asserted-by":"publisher","first-page":"155","DOI":"10.1016\/j.jcm.2016.02.012","article-title":"A guideline of selecting and reporting intraclass correlation coefficients for reliability research","volume":"15","author":"Koo","year":"2016","journal-title":"J Chiropr Med"},{"key":"2026072905282772900_cjag056-B24","first-page":"231","article-title":"Normal interpupillary distance values in an Iranian population","volume":"7","author":"Fesharaki","year":"2012","journal-title":"J Ophthalmic Vis Res"},{"key":"2026072905282772900_cjag056-B25","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1117\/12.529999","volume-title":"Stereoscopic Displays and Virtual Reality Systems XI","author":"Dodgson","year":"2004"},{"key":"2026072905282772900_cjag056-B26","doi-asserted-by":"publisher","first-page":"20900","DOI":"10.1038\/s41598-024-71262-0","article-title":"Effects of facial forms and midline deviations and rolls on female facial beauty according to laypeople, orthodontists, and prosthodontists or restorative dentists","volume":"14","author":"Rakhshan","year":"2024","journal-title":"Sci Rep"},{"key":"2026072905282772900_cjag056-B27","doi-asserted-by":"publisher","first-page":"3349","DOI":"10.1109\/TPAMI.2020.2983686","article-title":"Deep high-resolution representation learning for visual recognition","volume":"43","author":"Wang","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2026072905282772900_cjag056-B28","author":"Tan"},{"key":"2026072905282772900_cjag056-B29","author":"He"},{"key":"2026072905282772900_cjag056-B30","author":"Howard"}],"container-title":["European Journal of Orthodontics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/ejo\/article-pdf\/48\/4\/cjag056\/70333962\/cjag056.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/ejo\/article-pdf\/48\/4\/cjag056\/70333962\/cjag056.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T09:28:33Z","timestamp":1785317313000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/ejo\/article\/doi\/10.1093\/ejo\/cjag056\/8746623"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,2]]},"references-count":30,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,6,2]]}},"URL":"https:\/\/doi.org\/10.1093\/ejo\/cjag056","relation":{},"ISSN":["0141-5387","1460-2210"],"issn-type":[{"value":"0141-5387","type":"print"},{"value":"1460-2210","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,8]]},"published":{"date-parts":[[2026,6,2]]},"article-number":"cjag056"}}