{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T20:06:49Z","timestamp":1792181209740,"version":"4.4.0"},"reference-count":35,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2025,6,14]],"date-time":"2025-06-14T00:00:00Z","timestamp":1749859200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Aims: The aim of this study was to evaluate the effectiveness of large language model (LLM)-based chatbot systems in predicting bone age and identifying growth stages, and to explore their potential as practical, infrastructure-independent alternatives to conventional methods and convolutional neural network (CNN)-based deep learning models. Methods: This study evaluated the performance of three ChatGPT-based models (GPT-4o, GPT-o4-mini-high, and GPT-o1-pro) in predicting bone age and growth stage using 90 anonymized hand\u2013wrist radiographs (30 from each growth stage\u2014pre-peak, peak, and post-peak\u2014with equal male and female distribution). Reference standards were ensured by expert orthodontists using Fishman\u2019s Skeletal Maturity Indicators (SMI) system and the Greulich\u2013Pyle Atlas, with each radiograph analyzed by three GPT models using standardized prompts. Model performances were evaluated through statistical analyses assessing agreement and prediction accuracy. Results: All models showed significant agreement with the reference values in bone age prediction (p &lt; 0.001), with GPT-o1-pro having the highest concordance (Pearson r = 0.546). No statistically significant difference was observed in the mean absolute error (MAE) among the models (p &gt; 0.05). The GPT-o4-mini-high model achieved an accuracy rate of 72.2% within a \u00b12 year deviation range for bone age prediction. The GPT-o1-pro and GPT-o4-mini-high models showed bias in the Bland\u2013Altman analysis of bone age predictions; however, GPT-o1-pro yielded more reliable predictions with narrower limits of agreement. In terms of growth stage classification, the GPT-4o model achieved the highest agreement with the reference values (\u03ba = 0.283, p &lt; 0.001). Conclusions: This study shows that general-purpose GPT models can support bone age and growth stages prediction, with each model having distinct strengths. While GPT models do not replace clinical examination, their contextual reasoning and ability to perform preliminary assessments without domain-specific training make them promising tools, though further development is needed.<\/jats:p>","DOI":"10.3390\/diagnostics15121513","type":"journal-article","created":{"date-parts":[[2025,6,16]],"date-time":"2025-06-16T04:06:24Z","timestamp":1750046784000},"page":"1513","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":24,"title":["Can AI-Based ChatGPT Models Accurately Analyze Hand\u2013Wrist Radiographs? A Comparative Study"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-6804-1276","authenticated-orcid":false,"given":"Ahmet","family":"Y\u0131ld\u0131r\u0131m","sequence":"first","affiliation":[{"name":"Department of Orthodontics, Faculty of Dentistry, Zonguldak Bulent Ecevit University, Zonguldak 67600, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8172-6043","authenticated-orcid":false,"given":"Orhan","family":"Cicek","sequence":"additional","affiliation":[{"name":"Department of Orthodontics, Faculty of Dentistry, Zonguldak Bulent Ecevit University, Zonguldak 67600, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0556-2830","authenticated-orcid":false,"given":"Yavuz Selim","family":"Gen\u00e7","sequence":"additional","affiliation":[{"name":"Samsun Oral and Dental Health Hospital, Samsun Provincial Health Directorate, Samsun 55060, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,6,14]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1207","DOI":"10.1109\/TMI.2016.2535865","article-title":"Lung pattern classification for interstitial lung diseases using a deep convolutional neural network","volume":"35","author":"Anthimopoulos","year":"2016","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"903","DOI":"10.2319\/022019-127.1","article-title":"Automated identification of cephalometric landmarks: Part 1\u2014Comparisons between the latest deep-learning methods YOLOV3 and SSD","volume":"89","author":"Park","year":"2019","journal-title":"Angle Orthod."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"cjae017","DOI":"10.1093\/ejo\/cjae017","article-title":"Evidence-based potential of generative artificial intelligence large language models in orthodontics: A comparative study of ChatGPT, Google Bard, and Microsoft Bing","volume":"46","author":"Makrygiannakis","year":"2024","journal-title":"Eur. J. Orthod."},{"key":"ref_4","unstructured":"Adamopoulou, E., and Moussiades, L. (2020, January 5\u20137). An overview of chatbot technology. Proceedings of the Artificial Intelligence Applications and Innovations: 16th IFIP WG 12.5 International Conference, AIAI 2020, Neos Marmaras, Greece. Proceedings, Part II 16, 2020."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1098","DOI":"10.1111\/jerd.13046","article-title":"Implications of large language models such as ChatGPT for dental medicine","volume":"35","author":"Eggmann","year":"2023","journal-title":"J. Esthet. Restor. Dent."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Sallam, M. (2023). ChatGPT utility in healthcare education, research, and practice: Systematic review on the promising perspectives and valid concerns. Healthcare, 11.","DOI":"10.3390\/healthcare11060887"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"102744","DOI":"10.1016\/j.dsx.2023.102744","article-title":"ChatGPT: Is this version good for healthcare and research?","volume":"17","author":"Vaishya","year":"2023","journal-title":"Diabetes Metab. Syndr. Clin. Res. Rev."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"1911","DOI":"10.1002\/ksa.12639","article-title":"High identification and positive-negative discrimination but limited detailed grading accuracy of ChatGPT-4o in knee osteoarthritis radiographs","volume":"33","author":"Zhu","year":"2025","journal-title":"Knee Surg. Sports Traumatol. Arthrosc."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"545","DOI":"10.1111\/1754-9485.13261","article-title":"A review of medical image data augmentation techniques for deep learning applications","volume":"65","author":"Chlap","year":"2021","journal-title":"J. Med. Imaging Radiat. Oncol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1190","DOI":"10.1007\/s00167-023-07372-5","article-title":"A deeper dive into ChatGPT: History, use and future perspectives for orthopaedic research","volume":"31","author":"Ollivier","year":"2023","journal-title":"Knee Surg. Sports Traumatol. Arthrosc."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1007\/s10278-017-9955-8","article-title":"Fully automated deep learning system for bone age assessment","volume":"30","author":"Lee","year":"2017","journal-title":"J. Digit. Imaging"},{"key":"ref_12","first-page":"88","article-title":"Radiographic evaluation of skeletal maturation: A clinically oriented method based on hand-wrist films","volume":"52","author":"Fishman","year":"1982","journal-title":"Angle Orthod."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1374","DOI":"10.2214\/AJR.17.18224","article-title":"Computerized bone age estimation using deep learning based program: Evaluation of the accuracy and efficiency","volume":"209","author":"Kim","year":"2017","journal-title":"Am. J. Roentgenol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1515\/JPEM.1994.7.2.141","article-title":"Automatic bone age measurement using computerized image analysis","volume":"7","author":"Tanner","year":"1994","journal-title":"J. Pediatr. Endocrinol. Metab."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1109\/TMI.2008.926067","article-title":"The BoneXpert method for automated determination of skeletal maturity","volume":"28","author":"Thodberg","year":"2008","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"591","DOI":"10.1007\/s00247-008-1090-8","article-title":"Automatic determination of Greulich and Pyle bone age in healthy Dutch children","volume":"39","author":"Lequin","year":"2009","journal-title":"Pediatr. Radiol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"322","DOI":"10.1016\/j.patcog.2011.06.003","article-title":"Bone age cluster assessment and feature clustering analysis based on phalangeal image rough segmentation","volume":"45","author":"Lin","year":"2012","journal-title":"Pattern Recognit."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"283","DOI":"10.1007\/s10278-018-0135-2","article-title":"Accurate age determination for adolescents using magnetic resonance imaging of the hand and wrist with an artificial neural network-based approach","volume":"32","author":"Tang","year":"2019","journal-title":"J. Digit. Imaging"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1151","DOI":"10.3348\/kjr.2023.0092","article-title":"Bone age assessment using artificial intelligence in Korean pediatric population: A comparison of deep-learning models trained with healthy chronological and Greulich-Pyle ages as labels","volume":"24","author":"Kim","year":"2023","journal-title":"Korean J. Radiol."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Haak, D., Simon, H., Yu, J., Harmsen, M., and Deserno, T.M. (2013, January 3\u20135). Bone age assessment using support vector machine regression. Proceedings of the Bildverarbeitung f\u00fcr die Medizin 2013: Algorithmen-Systeme-Anwendungen, Heidelberg, Germany. Proceedings des Workshops vom 3. bis 5. M\u00e4rz 2013 in Heidelberg.","DOI":"10.1007\/978-3-642-36480-8_30"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"e253","DOI":"10.2196\/jmir.9364","article-title":"Relationship between internet health information and patient compliance based on trust: Empirical study","volume":"20","author":"Lu","year":"2018","journal-title":"J. Med. Internet Res."},{"key":"ref_22","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"2369","DOI":"10.1109\/TMI.2016.2546227","article-title":"Segmenting retinal blood vessels with deep neural networks","volume":"35","author":"Liskowski","year":"2016","journal-title":"IEEE Trans. Med. Imaging"},{"key":"ref_24","first-page":"614","article-title":"Biostatistics 104: Correlational analysis","volume":"44","author":"Chan","year":"2003","journal-title":"Singap. Med. J."},{"key":"ref_25","unstructured":"(2025, June 09). Hello GPT-4o. Available online: https:\/\/openai.com\/index\/hello-gpt-4o\/."},{"key":"ref_26","unstructured":"(2025, June 09). ChatGPT OpenAI o3 and o4-Mini Models FAQ (Enterprise & Edu Version). Available online: https:\/\/help.openai.com\/en\/articles\/9855712-chatgpt-openai-o3-and-o4-mini-models-faq-enterprise-edu-version."},{"key":"ref_27","unstructured":"(2025, June 09). Introducing ChatGPT Pro. Available online: https:\/\/openai.com\/index\/introducing-chatgpt-pro\/."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"597","DOI":"10.1016\/j.acra.2024.08.009","article-title":"Evaluating artificial intelligence competency in education: Performance of ChatGPT-4 in the American Registry of Radiologic Technologists (ARRT) radiography certification exam","volume":"32","author":"Halka","year":"2025","journal-title":"Acad. Radiol."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1001\/jamaophthalmol.2024.0017","article-title":"Accuracy of an artificial intelligence chatbot\u2019s interpretation of clinical ophthalmic images","volume":"142","author":"Mihalache","year":"2024","journal-title":"JAMA Ophthalmol."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1602","DOI":"10.21037\/qims-23-1180","article-title":"Assessing the role of GPT-4 in thyroid ultrasound diagnosis and treatment recommendations: Enhancing interpretability with a chain of thought approach","volume":"14","author":"Wang","year":"2024","journal-title":"Quant. Imaging Med. Surg."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"e232255","DOI":"10.1148\/radiol.232255","article-title":"Collaborative enhancement of consistency and accuracy in US diagnosis of thyroid nodules using large language models","volume":"310","author":"Wu","year":"2024","journal-title":"Radiology"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1041","DOI":"10.1007\/s12020-024-04066-x","article-title":"Assessing the feasibility of ChatGPT-4o and Claude 3-Opus in thyroid nodule classification based on ultrasound images","volume":"87","author":"Chen","year":"2024","journal-title":"Endocrine"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"222","DOI":"10.2340\/17453674.2021.1133","article-title":"A comparison of 3 different methods for assessment of skeletal age when treating leg-length discrepancies: An inter-and intra-observer study","volume":"93","author":"Breen","year":"2022","journal-title":"Acta Orthop."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1136\/adc.81.2.172","article-title":"Bone age assessment: A large scale comparison of the Greulich and Pyle, and Tanner and Whitehouse (TW2) methods","volume":"81","author":"Bull","year":"1999","journal-title":"Arch. Dis. Child."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1186\/s40164-024-00537-z","article-title":"Exploring whether ChatGPT-4 with image analysis capabilities can diagnose osteosarcoma from X-ray images","volume":"13","author":"Ren","year":"2024","journal-title":"Exp. Hematol. Oncol."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/15\/12\/1513\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T17:52:07Z","timestamp":1760032327000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/15\/12\/1513"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,14]]},"references-count":35,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2025,6]]}},"alternative-id":["diagnostics15121513"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics15121513","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,14]]}}}