{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T18:16:06Z","timestamp":1775844966406,"version":"3.50.1"},"reference-count":30,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,10,25]],"date-time":"2023-10-25T00:00:00Z","timestamp":1698192000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,10,25]],"date-time":"2023-10-25T00:00:00Z","timestamp":1698192000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Caries and molar-incisor hypomineralization (MIH) are among the most prevalent diseases worldwide and need to be reliably diagnosed. The use of dental photographs and artificial intelligence (AI) methods may potentially contribute to realizing accurate and automated diagnostic visual examinations in the future. Therefore, the present study aimed to develop an AI-based algorithm that can detect, classify and localize caries and MIH. This study included an image set of 18,179 anonymous photographs. Pixelwise image labeling was achieved by trained and calibrated annotators using the Computer Vision Annotation Tool (CVAT). All annotations were made according to standard methods and were independently checked by an experienced dentist. The entire image set was divided into training (<jats:italic>N<\/jats:italic>\u2009=\u200916,679), validation (<jats:italic>N<\/jats:italic>\u2009=\u2009500) and test sets (<jats:italic>N<\/jats:italic>\u2009=\u20091000). The AI-based algorithm was trained and finetuned over 250 epochs by using image augmentation and adapting a vision transformer network (SegFormer-B5). Statistics included the determination of the intersection over union (IoU), average precision (AP) and accuracy (ACC). The overall diagnostic performance in terms of IoU, AP and ACC were 0.959, 0.977 and 0.978 for the finetuned model, respectively. The corresponding data for the most relevant caries classes of non-cavitations (0.630, 0.813 and 0.990) and dentin cavities (0.692, 0.830, and 0.997) were found to be high. MIH-related demarcated opacity (0.672, 0.827, and 0.993) and atypical restoration (0.829, 0.902, and 0.999) showed similar results. Here, we report that the model achieves excellent precision for pixelwise detection and localization of caries and MIH. Nevertheless, the model needs to be further improved and externally validated.<\/jats:p>","DOI":"10.1038\/s41746-023-00944-2","type":"journal-article","created":{"date-parts":[[2023,10,25]],"date-time":"2023-10-25T20:34:45Z","timestamp":1698266085000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":40,"title":["Detection and localization of caries and hypomineralization on dental photographs with a vision transformer model"],"prefix":"10.1038","volume":"6","author":[{"given":"Marco","family":"Felsch","sequence":"first","affiliation":[]},{"given":"Ole","family":"Meyer","sequence":"additional","affiliation":[]},{"given":"Anne","family":"Schlickenrieder","sequence":"additional","affiliation":[]},{"given":"Paula","family":"Engels","sequence":"additional","affiliation":[]},{"given":"Jule","family":"Sch\u00f6newolf","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0007-0449-8458","authenticated-orcid":false,"given":"Felicitas","family":"Z\u00f6llner","sequence":"additional","affiliation":[]},{"given":"Roswitha","family":"Heinrich-Weltzien","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6826-9212","authenticated-orcid":false,"given":"Marc","family":"Hesenius","sequence":"additional","affiliation":[]},{"given":"Reinhard","family":"Hickel","sequence":"additional","affiliation":[]},{"given":"Volker","family":"Gruhn","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4063-2291","authenticated-orcid":false,"given":"Jan","family":"K\u00fchnisch","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,10,25]]},"reference":[{"key":"944_CR1","unstructured":"World Health Organization (WHO). Global Oral Health Status Report \u2013 Towards Universal Health Coverage for Oral Health by 2030 (WHO, 2022)."},{"key":"944_CR2","doi-asserted-by":"crossref","unstructured":"GBD 2017 Oral Disorders Collaborators et al. Global, regional, and national levels and trends in burden of oral conditions from 1990 to 2017: a systematic analysis for the Global Burden of Disease 2017 Study. J. Dent. Res. 99, 362\u2013373 (2020).","DOI":"10.1177\/0022034520908533"},{"key":"944_CR3","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.jdent.2017.12.002","volume":"68","author":"F Schwendicke","year":"2018","unstructured":"Schwendicke, F. et al. Global burden of molar incisor hypomineralization. J. Dent. 68, 10\u201318 (2018).","journal-title":"J. Dent."},{"key":"944_CR4","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1177\/00220345211042795","volume":"101","author":"T Walsh","year":"2022","unstructured":"Walsh, T. et al. Enamel caries detection and diagnosis: an analysis of systematic reviews. J. Dent. Res. 101, 261\u2013269 (2022).","journal-title":"J. Dent. Res."},{"key":"944_CR5","doi-asserted-by":"publisher","first-page":"6069","DOI":"10.1007\/s00784-021-04113-1","volume":"25","author":"M Janjic Rankovic","year":"2021","unstructured":"Janjic Rankovic, M. et al. Systematic review and meta-analysis of diagnostic studies of proximal surface caries. Clin. Oral Investig. 25, 6069\u20136079 (2021).","journal-title":"Clin. Oral Investig."},{"key":"944_CR6","doi-asserted-by":"publisher","first-page":"4801","DOI":"10.1007\/s00784-021-04024-1","volume":"25","author":"S Kapor","year":"2021","unstructured":"Kapor, S. et al. Systematic review and meta-analysis of diagnostic methods for occlusal surface caries. Clin. Oral Investig. 25, 4801\u20134815 (2021).","journal-title":"Clin. Oral Investig."},{"key":"944_CR7","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1111\/cdoe.12641","volume":"49","author":"T Gimenez","year":"2021","unstructured":"Gimenez, T. et al. What is the most accurate method for detecting caries lesions? A systematic review. Community Dent. Oral Epidemiol. 49, 216\u2013224 (2021).","journal-title":"Community Dent. Oral Epidemiol."},{"key":"944_CR8","doi-asserted-by":"publisher","first-page":"406","DOI":"10.1159\/000486429","volume":"52","author":"KR Ekstrand","year":"2018","unstructured":"Ekstrand, K. R., Gimenez, T., Ferreira, F. R., Mendes, F. M. & Braga, M. M. The International Caries Detection and Assessment System - ICDAS: a systematic review. Caries Res. 52, 406\u2013419 (2018).","journal-title":"Caries Res."},{"key":"944_CR9","doi-asserted-by":"publisher","first-page":"C67","DOI":"10.1177\/154405910408301s13","volume":"83","author":"KR Ekstrand","year":"2014","unstructured":"Ekstrand, K. R. Improving clinical visual detection-potential for caries clinical trials. J. Dent. Res. 83, C67\u2013C71 (2014).","journal-title":"J. Dent. Res."},{"key":"944_CR10","doi-asserted-by":"publisher","first-page":"158","DOI":"10.1177\/00220345211032524","volume":"101","author":"J K\u00fchnisch","year":"2022","unstructured":"K\u00fchnisch, J., Meyer, O., Hesenius, M., Hickel, R. & Gruhn, V. Caries detection on intraoral images using artificial intelligence. J. Dent. Res. 101, 158\u2013165 (2022).","journal-title":"J. Dent. Res."},{"key":"944_CR11","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1111\/odi.13735","volume":"28","author":"X Zhang","year":"2022","unstructured":"Zhang, X. et al. Development and evaluation of deep learning for screening dental caries from oral photographs. Oral Dis. 28, 173\u2013181 (2022).","journal-title":"Oral Dis."},{"key":"944_CR12","doi-asserted-by":"publisher","first-page":"103615","DOI":"10.1016\/j.jdent.2021.103615","volume":"107","author":"H Askar","year":"2021","unstructured":"Askar, H. et al. Detecting white spot lesions on dental photography using deep learning: a pilot study. J. Dent. 107, 103615 (2021).","journal-title":"J. Dent."},{"key":"944_CR13","doi-asserted-by":"publisher","first-page":"6917","DOI":"10.1007\/s00784-022-04646-z","volume":"26","author":"V Alevizakos","year":"2022","unstructured":"Alevizakos, V., Bekes, K., Steffen, R. & von See, C. Artificial intelligence system for training diagnosis and differentiation with molar incisor hypomineralization (MIH) and similar pathologies. Clin. Oral Investig. 26, 6917\u20136923 (2022).","journal-title":"Clin. Oral Investig."},{"key":"944_CR14","doi-asserted-by":"publisher","first-page":"5923","DOI":"10.1007\/s00784-022-04552-4","volume":"26","author":"J Sch\u00f6newolf","year":"2022","unstructured":"Sch\u00f6newolf, J. et al. Artificial intelligence-based diagnostics of molar-incisor-hypomineralization (MIH) on intraoral photographs. Clin. Oral Investig. 26, 5923\u20135930 (2022).","journal-title":"Clin. Oral Investig."},{"key":"944_CR15","unstructured":"Xie, E. et al. SegFormer: simple and efficient design for semantic segmentation with transformers. Preprint at https:\/\/arxiv.org\/abs\/2105.15203 (2021)."},{"key":"944_CR16","doi-asserted-by":"publisher","first-page":"257","DOI":"10.1007\/s11424-022-2057-9","volume":"36","author":"C Sheng","year":"2022","unstructured":"Sheng, C. et al. Transformer-based deep learning network for tooth segmentation on panoramic radiographs. J. Syst. Sci. Complex. 36, 257\u2013272 (2022).","journal-title":"J. Syst. Sci. Complex."},{"key":"944_CR17","doi-asserted-by":"publisher","first-page":"2625","DOI":"10.1016\/j.injury.2022.04.013","volume":"53","author":"L Tanzi","year":"2022","unstructured":"Tanzi, L., Audisio, A., Cirrincione, G., Aprato, A. & Vezzetti, E. Vision transformer for femur fracture classification. Injury 53, 2625\u20132634 (2022).","journal-title":"Injury"},{"key":"944_CR18","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-022-15634-4","volume":"12","author":"MN Islam","year":"2022","unstructured":"Islam, M. N. et al. Vision transformer and explainable transfer learning models for auto detection of kidney cyst, stone and tumor from CT-radiography. Sci. Rep. 12, 11440 (2022).","journal-title":"Sci. Rep."},{"key":"944_CR19","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-021-01456-3","volume":"21","author":"A Roshanzamir","year":"2021","unstructured":"Roshanzamir, A., Aghajan, H. & Soleymani Baghshah, M. Transformer-based deep neural network language models for Alzheimer\u2019s disease risk assessment from targeted speech. BMC Med. Inform. Decis. Mak. 21, 92 (2021).","journal-title":"BMC Med. Inform. Decis. Mak."},{"key":"944_CR20","doi-asserted-by":"publisher","first-page":"h5527","DOI":"10.1136\/bmj.h5527","volume":"351","author":"PM Bossuyt","year":"2015","unstructured":"Bossuyt, P. M. et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ 351, h5527 (2015).","journal-title":"BMJ"},{"key":"944_CR21","doi-asserted-by":"publisher","first-page":"103610","DOI":"10.1016\/j.jdent.2021.103610","volume":"107","author":"F Schwendicke","year":"2021","unstructured":"Schwendicke, F. et al. Artificial intelligence in dental research: checklist for authors, reviewers, readers. J. Dent. 107, 103610 (2021).","journal-title":"J. Dent."},{"key":"944_CR22","doi-asserted-by":"publisher","first-page":"104124","DOI":"10.1016\/j.jdent.2022.104124","volume":"121","author":"P Engels","year":"2022","unstructured":"Engels, P. et al. Automated detection of posterior restorations in permanent teeth using artificial intelligence on intraoral photographs. J. Dent. 121, 104124 (2022).","journal-title":"J. Dent."},{"key":"944_CR23","doi-asserted-by":"publisher","first-page":"1608","DOI":"10.3390\/diagnostics11091608","volume":"11","author":"A Schlickenrieder","year":"2021","unstructured":"Schlickenrieder, A. et al. Automatized detection and categorization of fissure sealants from intraoral digital photographs using artificial intelligence. Diagnostics 11, 1608 (2021).","journal-title":"Diagnostics"},{"key":"944_CR24","unstructured":"World Health Organization (WHO). Oral Health Surveys: Basic Methods 5th edn (WHO, 2013)."},{"key":"944_CR25","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1007\/s00784-010-0390-1","volume":"15","author":"J K\u00fchnisch","year":"2011","unstructured":"K\u00fchnisch, J. et al. Diagnostic performance of the universal visual scoring system (UniViSS) on occlusal surfaces. Clin. Oral Investig. 15, 215\u2013223 (2011).","journal-title":"Clin. Oral Investig."},{"key":"944_CR26","doi-asserted-by":"publisher","first-page":"2500","DOI":"10.3390\/ijerph6092500","volume":"6","author":"J K\u00fchnisch","year":"2009","unstructured":"K\u00fchnisch, J. et al. Development, methodology and potential of the new Universal Visual Scoring System (UniViSS) for caries detection and diagnosis. Int. J. Environ. Res. Public Health 6, 2500\u20132509 (2009).","journal-title":"Int. J. Environ. Res. Public Health"},{"key":"944_CR27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1159\/000224208","volume":"21","author":"NB Pitts","year":"2009","unstructured":"Pitts, N. B. How the detection, assessment, diagnosis and monitoring of caries integrate with personalized caries management. Monogr. Oral Sci. 21, 1\u201314 (2009).","journal-title":"Monogr. Oral Sci."},{"key":"944_CR28","doi-asserted-by":"publisher","first-page":"252","DOI":"10.1159\/000016526","volume":"33","author":"B Nyvad","year":"1999","unstructured":"Nyvad, B., Machiulskiene, V. & Baelum, V. Reliability of a new caries diagnostic system differentiating between active and inactive caries lesions. Caries Res. 33, 252\u2013260 (1999).","journal-title":"Caries Res."},{"key":"944_CR29","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1007\/BF03262716","volume":"11","author":"NA Lygidakis","year":"2010","unstructured":"Lygidakis, N. A. et al. Best clinical practice guidance for clinicians dealing with children presenting with molar-incisor-hypomineralisation (MIH): an EAPD policy document. Eur. Arch. Paediatr. Dent. 11, 75\u201381 (2010).","journal-title":"Eur. Arch. Paediatr. Dent."},{"key":"944_CR30","doi-asserted-by":"publisher","first-page":"1057","DOI":"10.1007\/s00068-022-02136-1","volume":"49","author":"J Prijs","year":"2023","unstructured":"Prijs, J. et al. Development and external validation of automated detection, classification, and localization of ankle fractures: inside the black box of a convolutional neural network (CNN). Eur. J. Trauma Emerg. Surg. 49, 1057\u20131069 (2023).","journal-title":"Eur. J. Trauma Emerg. Surg."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00944-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00944-2","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00944-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,19]],"date-time":"2023-11-19T02:21:28Z","timestamp":1700360488000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00944-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,10,25]]},"references-count":30,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["944"],"URL":"https:\/\/doi.org\/10.1038\/s41746-023-00944-2","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,10,25]]},"assertion":[{"value":"6 March 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 October 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 October 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"198"}}