{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,18]],"date-time":"2026-10-18T09:24:58Z","timestamp":1792315498808,"version":"4.4.0"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Universit\u00e4tsklinikum Bonn"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Orofac Orthop"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study aimed to evaluate the performance of large language models (LLMs), specifically GPT-4o and Gemini\u00a02 Flash, in identifying external apical root resorption (EARR) on panoramic radiographs of orthodontic patients using a\u00a0standardized prompt.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This comparative observational diagnostic study included 52\u00a0cropped tooth images obtained from panoramic radiographs of healthy individuals after orthodontic treatment. From each image, the regions corresponding to the permanent maxillary and mandibular incisors were manually cropped to include the apex, surrounding alveolar bone, and crown. An expert in endodontics evaluated each cropped image for the presence and severity of EARR using the Malmgren scale. The same images were submitted to both LLMs (GPT-4o and Gemini 2\u00a0Flash) using an identical multimodal prompt (image and text instructions). The models\u2019 responses were compared to the expert ratings using Cohen\u2019s kappa (\u03ba), accuracy, F1-score, mean absolute error (MAE), and confusion matrices. The Wilcoxon signed-rank test was used to compare the MAE between the models. Confidence intervals (95%) were calculated via bootstrapping.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      According to the expert evaluation, EARR was identified across all Malmgren grades: 11\u00a0teeth (21.2%) showed no resorption, 11\u00a0(21.2%) irregular apical contour, 11\u00a0(21.2%) small apical resorption, 11\u00a0(21.2%) resorption up to one-third of the root length, and 8\u00a0(15.4%) exceeding one-third resorption. GPT-4o showed fair agreement with the expert for the binary classification (\u03ba=\u20090.371), whereas Gemini exhibited a\u00a0negative \u03ba (\u22120.152), indicating performance below chance. GPT-4o achieved 36.5% accuracy and a\u00a0MAE of 1.269 in severity classification, compared to 13.5% accuracy and a\u00a0MAE of 1.750 for Gemini. Both models performed poorly in detecting moderate to severe EARR. The Wilcoxon test showed no significant difference between the models (\n                      <jats:italic>p<\/jats:italic>\n                      \u202f&gt;\u20090.05).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>GPT-4o achieved numerically better results than Gemini, with lower error rates and slightly higher agreement with the expert. Nevertheless, both models showed limited accuracy and agreement, particularly in detecting moderate to severe resorption, and neither can be considered suitable for clinical application at this stage.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00056-025-00624-8","type":"journal-article","created":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T14:47:29Z","timestamp":1765810049000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["Comparative evaluation of ChatGPT and Gemini in detecting external apical root resorption on panoramic radiographs of orthodontic patients","Vergleich von ChatGPT und Gemini zur Erkennung externer apikaler Wurzelresorption auf Panoramaschichtaufnahmen von kieferorthop\u00e4dischen Patienten"],"prefix":"10.1007","author":[{"given":"Allan","family":"Abuabara","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Flares","family":"Baratto-Filho","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Giancarlo Roos","family":"Gallego","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luana Beatriz das Portas","family":"Luiz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michelle Nascimento","family":"Meger","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rafaela","family":"Scariot","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Svenja","family":"Beisel-Memmert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cristiano Miranda","family":"de Araujo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erika Calvano","family":"K\u00fcchler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bianca Marques de Mattos","family":"de Araujo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,12,15]]},"reference":[{"key":"624_CR1","doi-asserted-by":"publisher","DOI":"10.1007\/s12024-025-01042-3","author":"A Abuabara","year":"2025","unstructured":"Abuabara\u00a0A, do Nascimento\u00a0TVPM, da Cruz\u00a0KR, K\u00fcchler\u00a0EC, Madalena\u00a0IR, de Oliveira\u00a0MBCR et\u00a0al (2025) Dental age estimation by comparing demirjian\u2019s method and machine learning in southeast brazilian youth. Forensic Sci Med Pathol. https:\/\/doi.org\/10.1007\/s12024-025-01042-3","journal-title":"Forensic Sci Med Pathol"},{"issue":"2","key":"624_CR2","doi-asserted-by":"publisher","first-page":"100","DOI":"10.34172\/joddd.2021.017","volume":"15","author":"F Bayir","year":"2021","unstructured":"Bayir\u00a0F, Bolat Gumus\u00a0E (2021) External apical root resorption after orthodontic treatment: incidence, severity and risk factors. J\u00a0Dent Res Dent Clin Dent Prospects 15(2):100\u2013105. https:\/\/doi.org\/10.34172\/joddd.2021.017","journal-title":"J Dent Res Dent Clin Dent Prospects"},{"issue":"25","key":"624_CR3","doi-asserted-by":"publisher","first-page":"00267","DOI":"10.1016\/j.joen.2025.05.012","volume":"S0099\u20132399","author":"DP Bubna","year":"2025","unstructured":"Bubna\u00a0DP, de Jesus Freitas\u00a0FP, Ferraz\u00a0AX, Abuabara\u00a0A, Baratto-Filho\u00a0F, Marques de Mattos de\u00a0AB et\u00a0al (2025) Dental trauma Evo\u2014development of an artificial intelligence-powered chatbot to support professional management of dental trauma. J\u00a0Endod S0099\u20132399(25):267\u2013265. https:\/\/doi.org\/10.1016\/j.joen.2025.05.012","journal-title":"J Endod"},{"issue":"4","key":"624_CR4","doi-asserted-by":"publisher","first-page":"356","DOI":"10.1016\/j.ajodo.2024.05.017","volume":"166","author":"JM Burnheimer","year":"2024","unstructured":"Burnheimer\u00a0JM, Baxter\u00a0DJ, Deeley\u00a0KB, Vieira\u00a0AR, Bezamat\u00a0M (2024) Exploring etiologic contributions to the occurrence of external apical root resorption. Am J Orthod Dentofacial Orthop 166(4):356\u2013362.e8. https:\/\/doi.org\/10.1016\/j.ajodo.2024.05.017","journal-title":"Am J Orthod Dentofacial Orthop"},{"issue":"12","key":"624_CR5","doi-asserted-by":"publisher","first-page":"1761","DOI":"10.1016\/j.joen.2024.09.008","volume":"50","author":"BMM de Araujo","year":"2024","unstructured":"de Araujo\u00a0BMM, de Jesus Freitas\u00a0PF, Deliga Schroder\u00a0AG, K\u00fcchler\u00a0EC, Baratto-Filho\u00a0F, Ditzel Westphalen\u00a0VP et\u00a0al (2024) PAIne: an artificial intelligence-based virtual assistant to aid in the differentiation of pain of odontogenic versus temporomandibular origin. J\u00a0Endod 50(12):1761\u20131765.e2. https:\/\/doi.org\/10.1016\/j.joen.2024.09.008","journal-title":"J Endod"},{"key":"624_CR6","doi-asserted-by":"publisher","first-page":"1624786","DOI":"10.3389\/fdgth.2025.1624786","volume":"7","author":"F Del Monte","year":"2025","unstructured":"Del Monte\u00a0F, Barolo\u00a0R, Circhetta\u00a0M, Delmonaco\u00a0AG, Castagno\u00a0E, Pivetta\u00a0E et\u00a0al (2025) Diagnostic efficacy of large language models in the pediatric emergency department: a\u00a0pilot study. Front Digit Health 7:1624786. https:\/\/doi.org\/10.3389\/fdgth.2025.1624786","journal-title":"Front Digit Health"},{"issue":"3","key":"624_CR7","doi-asserted-by":"publisher","first-page":"527","DOI":"10.1007\/s40368-025-01012-x","volume":"26","author":"A Dermata","year":"2025","unstructured":"Dermata\u00a0A, Arhakis\u00a0A, Makrygiannakis\u00a0MA, Giannakopoulos\u00a0K, Kaklamanos\u00a0EG (2025) Evaluating the evidence-based potential of six large language models in paediatric dentistry: a\u00a0comparative study on generative artificial intelligence. Eur Arch Paediatr Dent 26(3):527\u2013535. https:\/\/doi.org\/10.1007\/s40368-025-01012-x","journal-title":"Eur Arch Paediatr Dent"},{"issue":"2","key":"624_CR8","doi-asserted-by":"publisher","first-page":"102095","DOI":"10.1016\/j.jebdp.2025.102095","volume":"25","author":"NF Estrella","year":"2025","unstructured":"Estrella\u00a0NF, Alexandra\u00a0DS, Yun\u00a0C, Palma-Fern\u00e1ndez\u00a0JC, Alejandro\u00a0IL (2025) AI-aided volumetric root resorption assessment following personalized forces in orthodontics: preliminary results of a\u00a0randomized clinical trial. J\u00a0Evid Based Dent Pract 25(2):102095. https:\/\/doi.org\/10.1016\/j.jebdp.2025.102095","journal-title":"J Evid Based Dent Pract"},{"key":"624_CR9","doi-asserted-by":"publisher","first-page":"e545","DOI":"10.5114\/pjr.2022.120512","volume":"87","author":"K Futyma-G\u0105bka","year":"2022","unstructured":"Futyma-G\u0105bka\u00a0K, R\u00f3\u017cy\u0142o-Kalinowska\u00a0I, Pisk\u00f3rz\u00a0M, Bis\u00a0E, Borek\u00a0W (2022) Evaluation of root resorption in maxillary anterior teeth during orthodontic treatment with a\u00a0fixed appliance based on panoramic radiographs. Pol J Radiol 87:e545\u2013e548. https:\/\/doi.org\/10.5114\/pjr.2022.120512","journal-title":"Pol J Radiol"},{"key":"624_CR10","doi-asserted-by":"publisher","first-page":"e51580","DOI":"10.2196\/51580","volume":"25","author":"K Giannakopoulos","year":"2023","unstructured":"Giannakopoulos\u00a0K, Kavadella\u00a0A, Aaqel\u00a0SA, Stamatopoulos\u00a0V, Kaklamanos\u00a0EG (2023) Evaluation of the performance of generative AI large language models chatGPT, Google bard, and microsoft bing chat in supporting evidence-based dentistry: comparative mixed methods study. J\u00a0Med Internet Res 25:e51580. https:\/\/doi.org\/10.2196\/51580","journal-title":"J Med Internet Res"},{"issue":"3","key":"624_CR11","doi-asserted-by":"publisher","first-page":"368504221109217","DOI":"10.1177\/00368504221109217","volume":"105","author":"A Heboyan","year":"2022","unstructured":"Heboyan\u00a0A, Avetisyan\u00a0A, Karobari\u00a0MI, Marya\u00a0A, Khurshid\u00a0Z, Rokaya\u00a0D et\u00a0al (2022) Tooth root resorption: a\u00a0review. Sci Prog 105(3):368504221109217. https:\/\/doi.org\/10.1177\/00368504221109217","journal-title":"Sci Prog"},{"issue":"8","key":"624_CR12","doi-asserted-by":"publisher","first-page":"3872","DOI":"10.4103\/jfmpc.jfmpc_802_20","volume":"9","author":"S Kalra","year":"2020","unstructured":"Kalra\u00a0S, Gupta\u00a0P, Tripathi\u00a0T, Rai\u00a0P (2020) External apical root resorption in orthodontic patients: molecular and genetic basis. J\u00a0Family Med Prim Care 9(8):3872\u20133882. https:\/\/doi.org\/10.4103\/jfmpc.jfmpc_802_20","journal-title":"J Family Med Prim Care"},{"issue":"4","key":"624_CR13","first-page":"506","volume":"30","author":"AR Khan","year":"2018","unstructured":"Khan\u00a0AR, Fida\u00a0M, Shaikh\u00a0A (2018) Evaluation of apical root resorption in endodontically treated and vital teeth in adult orthodontic subjects. J\u00a0Ayub Med Coll Abbottabad 30(4):506\u2013510","journal-title":"J Ayub Med Coll Abbottabad"},{"key":"624_CR14","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1016\/j.jdent.2018.07.015","volume":"77","author":"JH Lee","year":"2018","unstructured":"Lee\u00a0JH, Kim\u00a0DH, Jeong\u00a0SN, Choi\u00a0SH (2018) Detection and diagnosis of dental caries using a\u00a0deep learning-based convolutional neural network algorithm. J\u00a0Dent 77:106\u2013111. https:\/\/doi.org\/10.1016\/j.jdent.2018.07.015","journal-title":"J Dent"},{"issue":"1","key":"624_CR15","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1093\/ejo\/10.1.30","volume":"10","author":"E Levander","year":"1988","unstructured":"Levander\u00a0E, Malmgren\u00a0O (1988) Evaluation of the risk of root resorption during orthodontic treatment: a\u00a0study of upper incisors. Eur J Orthod 10(1):30\u201338. https:\/\/doi.org\/10.1093\/ejo\/10.1.30","journal-title":"Eur J Orthod"},{"issue":"1","key":"624_CR16","doi-asserted-by":"publisher","first-page":"cjad059","DOI":"10.1093\/ejo\/cjad059","volume":"46","author":"RA Oster","year":"2024","unstructured":"Litchfield FE 4th, Oster\u00a0RA, Kau\u00a0CH, Lamani\u00a0E (2024) External apical root resorption in african american orthodontic patients. Eur J Orthod 46(1):cjad59. https:\/\/doi.org\/10.1093\/ejo\/cjad059","journal-title":"Eur J Orthod"},{"issue":"2","key":"624_CR17","doi-asserted-by":"publisher","first-page":"156","DOI":"10.12865\/chsj.45.02.05","volume":"45","author":"IR Marinescu","year":"2019","unstructured":"Marinescu\u00a0IR, B\u0103nic\u0103\u00a0AC, Mercu\u0163\u00a0V, Gheorghe\u00a0AG, Dr\u0103ghici\u00a0EC, Cojocaru\u00a0MO et\u00a0al (2019) Root resorption diagnostic: role of digital panoramic radiography. Curr Health Sci\u00a0J 45(2):156\u2013166. https:\/\/doi.org\/10.12865\/chsj.45.02.05","journal-title":"Curr Health Sci J"},{"issue":"1","key":"624_CR18","doi-asserted-by":"publisher","first-page":"648","DOI":"10.1186\/s12903-025-06050-x","volume":"25","author":"Y \u00d6zbay","year":"2025","unstructured":"\u00d6zbay\u00a0Y, Erdo\u011fan\u00a0D, Din\u00e7er\u00a0GA (2025) Evaluation of the performance of large language models in clinical decision-making in endodontics. BMC Oral Health 25(1):648. https:\/\/doi.org\/10.1186\/s12903-025-06050-x","journal-title":"BMC Oral Health"},{"key":"624_CR19","doi-asserted-by":"publisher","first-page":"792","DOI":"10.1111\/iej.13916","volume":"56","author":"S Patel","year":"2023","unstructured":"Patel\u00a0S, Krastl\u00a0G, Weiger\u00a0R, Lambrechts\u00a0P, Tj\u00e4derhane\u00a0L, Gambarini\u00a0G, Teng\u00a0PH (2023) ESE position statement on root resorption. Int Endod\u00a0J 56:792\u2013801. https:\/\/doi.org\/10.1111\/iej.13916","journal-title":"Int Endod J"},{"key":"624_CR20","doi-asserted-by":"publisher","first-page":"892","DOI":"10.1111\/iej.13715","volume":"55","author":"S Patel","year":"2022","unstructured":"Patel\u00a0S, Saberi\u00a0N, Pimental\u00a0T, Teng\u00a0PH (2022) Present status and future directions: root resorption. Int Endod\u00a0J 55:892\u2013921. https:\/\/doi.org\/10.1111\/iej.13715","journal-title":"Int Endod J"},{"issue":"1","key":"624_CR21","doi-asserted-by":"publisher","first-page":"252","DOI":"10.1186\/s12903-024-03910-w","volume":"24","author":"NH Reduwan","year":"2024","unstructured":"Reduwan\u00a0NH, Aziz\u00a0AA, Mohd\u00a0RR, Abdullah\u00a0ERMF, Mazloom Nezhad\u00a0SM, Gohain\u00a0M et\u00a0al (2024) Application of deep learning and feature selection technique on external root resorption identification on CBCT images. BMC Oral Health 24(1):252. https:\/\/doi.org\/10.1186\/s12903-024-03910-w","journal-title":"BMC Oral Health"},{"issue":"1","key":"624_CR22","doi-asserted-by":"publisher","first-page":"43","DOI":"10.3978\/j.issn.2223-4292.2013.02.07","volume":"3","author":"A Sabarudin","year":"2013","unstructured":"Sabarudin\u00a0A, Tiau\u00a0YJ (2013) Image quality assessment in panoramic dental radiography: a\u00a0comparative study between conventional and digital systems. Quant Imaging Med Surg 3(1):43\u201348. https:\/\/doi.org\/10.3978\/j.issn.2223-4292.2013.02.07","journal-title":"Quant Imaging Med Surg"},{"issue":"2","key":"624_CR23","doi-asserted-by":"publisher","first-page":"220","DOI":"10.1016\/j.ajodo.2025.02.017","volume":"168","author":"F Salmanpour","year":"2025","unstructured":"Salmanpour\u00a0F, Akp\u0131nar\u00a0M (2025) Performance of chat generative pretrained transformer\u20114.0 in determining labiolingual localization of maxillary impacted canine and presence of resorption in incisors through panoramic radiographs: a\u00a0retrospective study. Am J Orthod Dentofacial Orthop 168(2):220\u2013231","journal-title":"Am J Orthod Dentofacial Orthop"},{"issue":"3","key":"624_CR24","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1007\/s00784-024-05598-2","volume":"28","author":"Y Shi","year":"2024","unstructured":"Shi\u00a0Y, Ye\u00a0Z, Guo\u00a0J, Tang\u00a0Y, Dong\u00a0W, Dai\u00a0J et\u00a0al (2024) Deep learning methods for fully automated dental age estimation on orthopantomograms. Clin Oral Investig 28(3):198. https:\/\/doi.org\/10.1007\/s00784-024-05598-2","journal-title":"Clin Oral Investig"},{"issue":"6","key":"624_CR25","doi-asserted-by":"publisher","first-page":"619","DOI":"10.4103\/0970-9290.176927","volume":"26","author":"WM Takeshita","year":"2015","unstructured":"Takeshita\u00a0WM, Chicarelli\u00a0M, Iwaki\u00a0LC (2015) Comparison of diagnostic accuracy of root perforation, external resorption and fractures using cone-beam computed tomography, panoramic radiography and conventional & digital periapical radiography. Indian J Dent Res 26(6):619\u2013626. https:\/\/doi.org\/10.4103\/0970-9290.176927","journal-title":"Indian J Dent Res"},{"issue":"8","key":"624_CR26","doi-asserted-by":"publisher","first-page":"e68085","DOI":"10.7759\/cureus.68085","volume":"16","author":"S Vassis","year":"2024","unstructured":"Vassis\u00a0S, Powell\u00a0H, Petersen\u00a0E, Barkmann\u00a0A, Noeldeke\u00a0B, Kristensen\u00a0KD et\u00a0al (2024) Large-language models in orthodontics: assessing reliability and validity of chatGPT in pretreatment patient education. Cureus 16(8):e68085. https:\/\/doi.org\/10.7759\/cureus.68085","journal-title":"Cureus"}],"container-title":["Journal of Orofacial Orthopedics \/ Fortschritte der Kieferorthop\u00e4die"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00056-025-00624-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00056-025-00624-8\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00056-025-00624-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T14:47:33Z","timestamp":1765810053000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00056-025-00624-8"}},"subtitle":["Artificial intelligence performance in detecting root resorption","Performance von k\u00fcnstlicher Intelligenz bei der Erkennung von Wurzelresorption"],"short-title":[],"issued":{"date-parts":[[2025,12,15]]},"references-count":26,"alternative-id":["624"],"URL":"https:\/\/doi.org\/10.1007\/s00056-025-00624-8","relation":{},"ISSN":["1434-5293","1615-6714"],"issn-type":[{"value":"1434-5293","type":"print"},{"value":"1615-6714","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,15]]},"assertion":[{"value":"24 July 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 October 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"15 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"A. Abuabara, F. Baratto-Filho, G. Gallego, L. Luiz, M. Meger, R. Scariot, S. Beisel-Memmert, C.M. de Araujo, E. K\u00fcchler and B. de Araujo declare that there are no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Written informed consent was obtained from all patients prior to participation. The study was approved by the Ethics Committee of Tuiuti University of Paran\u00e1 under protocol number 6.973.530 and was conducted in accordance with the principles of the Declaration of Helsinki.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical standards"}}]}}