{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T09:38:38Z","timestamp":1793353118648,"version":"4.5.0"},"reference-count":23,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:00:00Z","timestamp":1791417600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T00:00:00Z","timestamp":1792195200000},"content-version":"vor","delay-in-days":9,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Oral Health"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Pulp stones are calcified structures within the dental pulp that may increase the complexity of endodontic treatment and are commonly identified through radiographic examinations. Despite recent advances in artificial intelligence applied to dental imaging, the diagnostic capability of publicly available multimodal large language models (LLMs) for detecting pulp stones under zero-shot conditions remains unclear. Therefore, this study evaluated the diagnostic performance and response consistency of multimodal LLMs in detecting pulp stones on panoramic radiographs of posterior teeth.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This comparative and exploratory study evaluated the GPT-5 and Gemini 2.5 Flash multimodal LLMs using panoramic radiograph image crops containing posterior teeth. A total of 64 image crops, comprising 192 posterior teeth, were analyzed. The models were tested under standardized zero-shot conditions using independent sessions and a standardized English-language prompt. Each image was submitted six times to both models to assess response consistency. Diagnostic performance was evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1-score with 95% confidence intervals. Intra-model consistency was assessed using majority agreement, complete agreement across repetitions, and Fleiss\u2019 Kappa coefficient.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The models demonstrated similar overall accuracy for binary detection of pulp stones, with 56.5% for ChatGPT and 57.0% for Gemini. ChatGPT showed high specificity (96.2%) and greater intra-model stability, whereas Gemini demonstrated higher sensitivity (81.8%) and superior performance in identifying the affected tooth and dental group. However, Gemini also presented lower specificity and greater variability across repeated evaluations. Fleiss\u2019 Kappa coefficients indicated low intra-model agreement beyond chance for both models, particularly for Gemini.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Multimodal LLMs demonstrated limited capability in detecting pulp stones on panoramic radiographs, with distinct diagnostic profiles between the evaluated models. Although the findings suggest potential future applications of LLMs as auxiliary tools for dental image interpretation, their current performance does not yet support consistent clinical applicability for pulp stone detection.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12903-026-10119-6","type":"journal-article","created":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T03:36:54Z","timestamp":1791430614000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Diagnostic performance of multimodal large language models in detecting pulp stones on panoramic radiographs"],"prefix":"10.1186","volume":"26","author":[{"given":"Bianca Marques de Mattos","family":"de Araujo","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria Eduarda","family":"Cavassin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Allan","family":"Abuabara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maria Luisa","family":"Giacon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luiz Fernando","family":"Fariniuk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christian","family":"Kirschneck","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":"Flares","family":"Baratto-Filho","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Erika Calvano","family":"K\u00fcchler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,10,8]]},"reference":[{"issue":"2","key":"10119_CR1","doi-asserted-by":"publisher","first-page":"192","DOI":"10.4103\/jomfp.jomfp_536_23","volume":"28","author":"MI Khan","year":"2024","unstructured":"Khan MI, Khare A, Arif K, Khan SS, Nasir A, Lari S. Dental pulp stones and their correlation with metabolic diseases. J oral maxillofacial pathology: JOMFP. 2024;28(2):192\u20139.","journal-title":"J oral maxillofacial pathology: JOMFP"},{"issue":"1","key":"10119_CR2","doi-asserted-by":"publisher","first-page":"1050","DOI":"10.1186\/s12903-025-06470-9","volume":"25","author":"N Ba\u011fc\u0131","year":"2025","unstructured":"Ba\u011fc\u0131 N, Yakubo\u011flu SK, Pamuk\u00e7u U, Toraman M. Association between pulp stone and Mediterranean diet: a clinical and radiographic study. BMC Oral Health. 2025;25(1):1050.","journal-title":"BMC Oral Health"},{"issue":"6","key":"10119_CR3","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1111\/j.1365-2591.2008.01374.x","volume":"41","author":"R Goga","year":"2008","unstructured":"Goga R, Chandler NP, Oginni AO. Pulp stones: a review. Int Endod J. 2008;41(6):457\u201368.","journal-title":"Int Endod J"},{"key":"10119_CR4","doi-asserted-by":"publisher","first-page":"605996","DOI":"10.3389\/fcell.2020.605996","volume":"8","author":"H Maeda","year":"2020","unstructured":"Maeda H. Aging and Senescence of Dental Pulp and Hard Tissues of the Tooth. Front Cell Dev Biology. 2020;8:605996.","journal-title":"Front Cell Dev Biology"},{"key":"10119_CR5","doi-asserted-by":"publisher","first-page":"7381610","DOI":"10.1155\/2023\/7381610","volume":"2023","author":"N Babanouri","year":"2023","unstructured":"Babanouri N, Sahmeddini S, Khoshmakani MR. Effects of Orthodontic Treatment on Pulp Stone Formation: A Retrospective Study. Biomed Res Int. 2023;2023:7381610.","journal-title":"Biomed Res Int"},{"issue":"4","key":"10119_CR6","doi-asserted-by":"publisher","first-page":"405","DOI":"10.2319\/091824-763.1","volume":"95","author":"R Ba-Hattab","year":"2025","unstructured":"Ba-Hattab R, Almashraqi AA, Nasrawi YH, Sunna S, Abu Alhaija ES. The effect of clear aligner and fixed orthodontic treatment on the development of pulp stones: a retrospective observational study. Angle Orthod. 2025;95(4):405\u201311.","journal-title":"Angle Orthod"},{"issue":"4","key":"10119_CR7","doi-asserted-by":"publisher","first-page":"e70181","DOI":"10.1002\/cre2.70181","volume":"11","author":"K Gholinezhad","year":"2025","unstructured":"Gholinezhad K, Ghorbani H, Seyedmajidi S, Sheikhzadeh S, Rahmati Kamel M. Pulp Stone Formation Following Fixed Orthodontic Treatment: A Panoramic Radiographic Comparison of Extraction and Non-Extraction Approaches. Clin Exp Dent Res. 2025;11(4):e70181.","journal-title":"Clin Exp Dent Res"},{"key":"10119_CR8","doi-asserted-by":"publisher","first-page":"2703","DOI":"10.3390\/jcm13092703","volume":"13","author":"A Chaniotis","year":"2024","unstructured":"Chaniotis A, Sousa Dias H, Chanioti A. Negotiation of Calcified Canals. J Clin Med vol. 2024;13:2703.","journal-title":"J Clin Med vol"},{"issue":"1","key":"10119_CR9","doi-asserted-by":"publisher","first-page":"1132","DOI":"10.1186\/s12903-024-04922-2","volume":"24","author":"L Ye","year":"2024","unstructured":"Ye L, Li S, Li C, Wang C, Wei X, Zhou W, Du Y. Pulp calcification identification on cone beam computed tomography: an artificial intelligence pilot study. BMC Oral Health. 2024;24(1):1132.","journal-title":"BMC Oral Health"},{"key":"10119_CR10","doi-asserted-by":"publisher","first-page":"e256861","DOI":"10.1590\/0103-644020256861","volume":"36","author":"DP Bubna","year":"2026","unstructured":"Bubna DP, Mattos NHR, Luiz L, Baratto-Filho F, Mattos-Calil MT, Silva-Sousa YTC, K\u00fcchler EC, Schroder \u00c2GD, Araujo CM, Araujo BMM. Can Large Language Models Detect Periapical Lesions in Anterior Teeth? A Comparative Study. Braz Dent J. 2026;36:e256861.","journal-title":"Braz Dent J"},{"key":"10119_CR11","doi-asserted-by":"publisher","unstructured":"Abuabara A, Baratto-Filho F, Gallego GR, Luiz LBDP, Meger MN, Scariot R, Beisel-Memmert S, de Araujo CM, K\u00fcchler EC, de Araujo BMM. Comparative evaluation of ChatGPT and Gemini in detecting external apical root resorption on panoramic radiographs of orthodontic patients: Artificial intelligence performance in detecting root resorption. J Orofac Orthop. 2025. https:\/\/doi.org\/10.1007\/s00056-025-00624-8.","DOI":"10.1007\/s00056-025-00624-8"},{"issue":"1","key":"10119_CR12","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1038\/s41591-018-0316-z","volume":"25","author":"A Esteva","year":"2019","unstructured":"Esteva A, Robicquet A, Ramsundar B, Kuleshov V, DePristo M, Chou K, Cui C, Corrado G, Thrun S, Dean J. A guide to deep learning in healthcare. Nat Med. 2019;25(1):24\u20139.","journal-title":"Nat Med"},{"key":"10119_CR13","doi-asserted-by":"publisher","first-page":"e47564","DOI":"10.2196\/47564","volume":"10","author":"Y Shahsavar","year":"2023","unstructured":"Shahsavar Y, Choudhury A. User Intentions to Use ChatGPT for Self-Diagnosis and Health-Related Purposes: Cross-sectional Survey Study. JMIR Hum Factors. 2023;10:e47564.","journal-title":"JMIR Hum Factors"},{"key":"10119_CR14","first-page":"311","volume":"24","author":"Q Gao","year":"2026","unstructured":"Gao Q, Li G. Computed Tomographic Analysis of Associations Between Pulp Stones and Demographic and Tooth-Related Factors in a North China Population. Oral Health Prev Dent. 2026;24:311\u20137.","journal-title":"Oral Health Prev Dent"},{"issue":"2","key":"10119_CR15","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1007\/s11282-025-00804-7","volume":"41","author":"C Gurhan","year":"2025","unstructured":"Gurhan C, Yigit H, Yilmaz S, Cetinkaya C. A novel deep learning-based pipeline architecture for pulp stone detection on panoramic radiographs. Oral Radiol. 2025;41(2):285\u201395.","journal-title":"Oral Radiol"},{"issue":"1","key":"10119_CR16","doi-asserted-by":"publisher","first-page":"2664903","DOI":"10.1080\/07853890.2026.2664903","volume":"58","author":"AA Madfa","year":"2026","unstructured":"Madfa AA, Alshammari AF, Anazi BA. Assessing diagnostic performance of multimodal AI and human experts in oral and maxillofacial radiography: a comparative analysis of ChatGPT, Grok, and MANUS. Ann Med. 2026;58(1):2664903.","journal-title":"Ann Med"},{"key":"10119_CR17","first-page":"22","volume":"18","author":"E Afsari","year":"2021","unstructured":"Afsari E, Niksolat E, Ostovar F, Karimi S. Comparison of Abundance of Premolar and Molar Pulp Stones before and after Orthodontic Treatment Using Panoramic Radiography. Front dentistry. 2021;18:22.","journal-title":"Front dentistry"},{"key":"10119_CR18","doi-asserted-by":"publisher","unstructured":"Alt\u0131nda\u011f A, Bahrilli S, \u00c7elik \u00d6, Bayrakdar \u0130\u015e, Orhan K. The Detection of Pulp Stones with Automatic Deep Learning in Panoramic Radiographies: An AI Pilot Study. Diagnostics (Basel). 2024;14(9):890. https:\/\/doi.org\/10.3390\/diagnostics14090890.","DOI":"10.3390\/diagnostics14090890"},{"issue":"3","key":"10119_CR19","doi-asserted-by":"publisher","first-page":"e0001306","DOI":"10.1371\/journal.pdig.0001306","volume":"5","author":"V Mavrych","year":"2026","unstructured":"Mavrych V, Yousef EM, Yaqinuddin A, Shaikh AA, Bolgova O. Visual recognition limitations in multimodal large language models: A comparative analysis of histological image interpretation. PLOS Digit Health. 2026;5(3):e0001306.","journal-title":"PLOS Digit Health"},{"key":"10119_CR20","doi-asserted-by":"publisher","first-page":"1644093","DOI":"10.3389\/frai.2025.1644093","volume":"8","author":"A ElSayed","year":"2025","unstructured":"ElSayed A, Updegrove GF. Limitations of broadly trained LLMs in interpreting orthopedic Walch glenoid classifications. Front Artif Intell. 2025;8:1644093.","journal-title":"Front Artif Intell"},{"key":"10119_CR21","doi-asserted-by":"publisher","first-page":"1674835","DOI":"10.3389\/fdgth.2025.1674835","volume":"7","author":"D Khovanova","year":"2025","unstructured":"Khovanova D, Vasilev Y, Vladzymyrskyy A, Omelyanskaya O, Pamova A, Arzamasov K. A comparative accuracy study of multimodal LLMs, VLM and agent-based framework for pulmonary nodule detection on chest radiographs. Front Digit Health. 2025;7:1674835.","journal-title":"Front Digit Health"},{"key":"10119_CR22","doi-asserted-by":"publisher","first-page":"104931","DOI":"10.1016\/j.jbi.2025.104931","volume":"171","author":"TY Chang","year":"2025","unstructured":"Chang TY, Gou Q, Zhao L, Zhou T, Chen H, Yang D, Ju H, Smith KE, Sun C, Pan J, Huang Y, He X, Zhang X, Xu D, Xu J, Bian J, Chen A. From image to report: automating lung cancer screening interpretation and reporting with vision-language models. J Biomed Inf. 2025;171:104931.","journal-title":"J Biomed Inf"},{"issue":"1","key":"10119_CR23","doi-asserted-by":"publisher","first-page":"2002","DOI":"10.1186\/s12903-025-07360-w","volume":"25","author":"RB Peker","year":"2025","unstructured":"Peker RB. Artificial intelligence versus human expertise: reliability of ChatGPT and the London atlas for dental age estimation using panoramic radiographs. BMC Oral Health. 2025;25(1):2002.","journal-title":"BMC Oral Health"}],"container-title":["BMC Oral Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12903-026-10119-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12903-026-10119-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12903-026-10119-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,17]],"date-time":"2026-10-17T20:27:46Z","timestamp":1792268866000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12903-026-10119-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,8]]},"references-count":23,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["10119"],"URL":"https:\/\/doi.org\/10.1186\/s12903-026-10119-6","relation":{},"ISSN":["1472-6831"],"issn-type":[{"value":"1472-6831","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,8]]},"assertion":[{"value":"20 May 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 October 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 October 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The study was approved by the Research Ethics Committee under protocol number 6.973.530 and conducted in accordance with the principles of the Declaration of Helsinki. All participants provided written informed consent.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"2043"}}