{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T14:49:47Z","timestamp":1785595787917,"version":"3.56.0"},"reference-count":18,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"7","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,7]]},"abstract":"\n OBJECTIVES:<\/jats:title>\n To evaluate and compare the ability of three popular open-source artificial intelligence platforms to diagnose common trauma-related fractures using radiologic imaging.<\/jats:p>\n <\/jats:sec>\n \n METHODS:<\/jats:title>\n <\/jats:sec>\n \n Design:<\/jats:title>\n Retrospective diagnostic performance comparison study.<\/jats:p>\n <\/jats:sec>\n \n Setting:<\/jats:title>\n Publicly accessible online radiologic imaging databases.<\/jats:p>\n <\/jats:sec>\n \n Patient Selection Criteria:<\/jats:title>\n Five common orthopedic trauma fractures were assessed: ankle, tibial plateau, intertrochanteric, femoral neck, and humerus. Radiographs and computed tomography (CT) images were collected. Images were randomly selected from confirmed diagnoses on Radiopaedia.org.<\/jats:p>\n <\/jats:sec>\n \n Outcome Measures and Comparisons:<\/jats:title>\n ChatGPT 5, Grok 3, and Claude 4.5 Sonnet were queried with each image. Diagnostic accuracy, sensitivity, specificity, positive and negative predictive values, and performance by modality (X-ray vs. CT) were assessed. The reference standard was the expert-verified diagnosis provided by Radiopaedia.org, limited to cases labeled with a \u201cdiagnosis certain\u201d tag.<\/jats:p>\n <\/jats:sec>\n \n RESULTS:<\/jats:title>\n \n Each model was provided with 30 radiographs and 20 CT images whenever possible. ChatGPT 5, Grok 3, and Claude 4.5 Sonnet accurately diagnosed diseased images in 26.8%, 18.8%, and 22.4% of cases, respectively. By fracture type, ChatGPT 5 demonstrated the highest correct classification rates for ankle (10%), femoral neck (38%), humerus (40%), and tibial plateau (44%) fractures. Grok 3 demonstrated the highest correct classification rate for intertrochanteric fractures (6%). Overall sensitivities were 0.267, 0.187, and 0.223 for ChatGPT 5, Grok 3, and Claude 4.5 Sonnet, respectively. ChatGPT 5 and Grok 3 outperformed Claude 4.5 Sonnet (both\n P<\/jats:italic>\n < 0.001). No modality-based performance differences were observed for any model.\n <\/jats:p>\n <\/jats:sec>\n \n CONCLUSIONS:<\/jats:title>\n Among the publicly available large language models evaluated for radiologic interpretation of orthopedic trauma imaging, ChatGPT 5 demonstrated the highest overall diagnostic accuracy, followed by Claude 4.5 Sonnet and Grok 3. Despite relative variation between the models, overall diagnostic accuracy for fracture detection was low across all platforms (<27%). In their baseline forms, these publicly accessible large language models are not recommended for radiologic imaging interpretation.<\/jats:p>\n <\/jats:sec>\n \n LEVEL OF EVIDENCE:<\/jats:title>\n Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.<\/jats:p>\n <\/jats:sec>","DOI":"10.1097\/bot.0000000000003166","type":"journal-article","created":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T21:02:57Z","timestamp":1772744577000},"page":"360-365","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":4,"title":["Comparing the Efficacy Between ChatGPT 5, Grok 3, and Claude 4.5 Sonnet in Analyzing Orthopedic Trauma-Related Imaging"],"prefix":"10.1097","volume":"40","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-8861-0594","authenticated-orcid":false,"given":"Joshua A.","family":"Holmstrom","sequence":"first","affiliation":[{"name":"Alix School of Medicine, Mayo Clinic, Scottsdale, AZ;"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4228-6154","authenticated-orcid":false,"given":"Collin L.","family":"Braithwaite","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, Mayo Clinic, Phoenix, AZ; and"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0311-1588","authenticated-orcid":false,"given":"Ahmad R.","family":"Alhankawi","sequence":"additional","affiliation":[{"name":"Alix School of Medicine, Mayo Clinic, Scottsdale, AZ;"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3003-3316","authenticated-orcid":false,"given":"M. Lane","family":"Moore","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, Mayo Clinic, Phoenix, AZ; and"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1107-5094","authenticated-orcid":false,"given":"Karan A.","family":"Patel","sequence":"additional","affiliation":[{"name":"Department of Orthopedic Surgery, Mayo Clinic, Phoenix, AZ; and"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brian H.","family":"Miller","sequence":"additional","affiliation":[{"name":"Sonoran Orthopaedic Trauma Surgeons, Scottsdale, AZ."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","reference":[{"key":"R1-20260612","doi-asserted-by":"crossref","first-page":"1740","DOI":"10.3390\/healthcare9121740","article-title":"Online health information seeking behavior: a systematic review","volume":"9","author":"Jia","year":"2021","journal-title":"Healthcare"},{"key":"R2-20260612","doi-asserted-by":"crossref","first-page":"581","DOI":"10.1080\/17453674.2017.1344459","article-title":"Artificial intelligence for analyzing orthopedic trauma radiographs","volume":"88","author":"Olczak","year":"2017","journal-title":"Acta Orthop."},{"key":"R3-20260612","doi-asserted-by":"crossref","first-page":"627","DOI":"10.1148\/radiol.210937","article-title":"Improving radiographic fracture recognition performance and efficiency using artificial intelligence","volume":"302","author":"Guermazi","year":"2022","journal-title":"Radiology"},{"key":"R4-20260612","doi-asserted-by":"crossref","first-page":"120","DOI":"10.1148\/radiol.2021203886","article-title":"Assessment of an AI aid in detection of adult appendicular skeletal fractures by emergency physicians and radiologists: a multicenter cross-sectional diagnostic study","volume":"300","author":"Duron","year":"2021","journal-title":"Radiology"},{"key":"R5-20260612","doi-asserted-by":"crossref","first-page":"23053","DOI":"10.1038\/s41598-024-73058-8","article-title":"Artificial intelligence in commercial fracture detection products: a systematic review and meta-analysis of diagnostic test accuracy","volume":"14","author":"Husarek","year":"2024","journal-title":"Sci Rep."},{"key":"R6-20260612","doi-asserted-by":"crossref","first-page":"e233391","DOI":"10.1001\/jamanetworkopen.2023.3391","article-title":"Artificial intelligence for hip fracture detection and outcome prediction: a systematic review and meta-analysis","volume":"6","author":"Lex","year":"2023","journal-title":"JAMA Netw Open."},{"key":"R7-20260612","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1016\/j.jhsa.2024.01.020","article-title":"Diagnostic performance of artificial intelligence for detection of scaphoid and distal radius fractures: a systematic review","volume":"49","author":"Oeding","year":"2024","journal-title":"J Hand Surg."},{"key":"R8-20260612","doi-asserted-by":"crossref","first-page":"110447","DOI":"10.1016\/j.ejrad.2022.110447","article-title":"Assessment of performances of a deep learning algorithm for the detection of limbs and pelvic fractures, dislocations, focal bone lesions, and elbow effusions on trauma X-rays","volume":"154","author":"Regnard","year":"2022","journal-title":"Eur J Radiol."},{"key":"R9-20260612","doi-asserted-by":"crossref","first-page":"2118","DOI":"10.1016\/j.acra.2023.06.016","article-title":"Artificial intelligence for detecting acute fractures in patients admitted to an emergency department: real-life performance of three commercial algorithms","volume":"30","author":"Bousson","year":"2023","journal-title":"Acad Radiol."},{"key":"R12-20260612","doi-asserted-by":"crossref","first-page":"987","DOI":"10.1016\/j.tics.2023.08.006","article-title":"What are large language models supposed to model?","volume":"27","author":"Blank","year":"2023","journal-title":"Trends Cogn Sci."},{"key":"R13-20260612","doi-asserted-by":"crossref","first-page":"e55627","DOI":"10.2196\/55627","article-title":"Evaluating ChatGPT-4's diagnostic accuracy: impact of visual data integration","volume":"12","author":"Hirosawa","year":"2024","journal-title":"JMIR Med Inf."},{"key":"R14-20260612","doi-asserted-by":"crossref","first-page":"e0164859","DOI":"10.1371\/journal.pone.0164859","article-title":"Comparison of diagnostic accuracy of radiation dose-equivalent radiography, multidetector computed tomography and cone beam computed tomography for fractures of adult cadaveric wrists","volume":"11","author":"Neubauer","year":"2016","journal-title":"PLoS One"},{"key":"R15-20260612","doi-asserted-by":"crossref","first-page":"9533573","DOI":"10.1155\/2021\/9533573","article-title":"Multislice spiral computed tomography postprocessing technology in the imaging diagnosis of extremities and joints","volume":"2021","author":"Yang","year":"2021","journal-title":"Comput Math Methods Med."},{"key":"R16-20260612","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1186\/s40634-023-00642-8","article-title":"Artificial intelligence and ChatGPT in orthopaedics and sports medicine","volume":"10","author":"Fayed","year":"2023","journal-title":"J Exp Orthop."},{"key":"R17-20260612","doi-asserted-by":"crossref","first-page":"50","DOI":"10.1148\/radiol.211785","article-title":"Artificial intelligence in fracture detection: a systematic review and meta-analysis","volume":"304","author":"Kuo","year":"2022","journal-title":"Radiology"},{"key":"R18-20260612","doi-asserted-by":"crossref","first-page":"7196","DOI":"10.1007\/s00330-022-08956-4","article-title":"Diagnostic accuracy and potential covariates of artificial intelligence for diagnosing orthopedic fractures: a systematic literature review and meta-analysis","volume":"32","author":"Zhang","year":"2022","journal-title":"Eur Radiol."},{"issue":"1","key":"R19-20260612","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1177\/15589447241298982","article-title":"Is artificial intelligence the future of radiology? Accuracy of ChatGPT in radiologic diagnosis of upper extremity bony pathology","volume":"21","author":"Hiredesai","year":"2024","journal-title":"Hand (N Y)"},{"key":"R21-20260612","doi-asserted-by":"crossref","first-page":"e210064","DOI":"10.1148\/ryai.210064","article-title":"External validation of deep learning algorithms for radiologic diagnosis: a systematic review","volume":"4","author":"Yu","year":"2022","journal-title":"Radiol Artif Intell."}],"container-title":["Journal of Orthopaedic Trauma"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.lww.com\/10.1097\/BOT.0000000000003166","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T13:00:11Z","timestamp":1781269211000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.lww.com\/10.1097\/BOT.0000000000003166"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":18,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.1097\/bot.0000000000003166","relation":{},"ISSN":["0890-5339","1531-2291"],"issn-type":[{"value":"0890-5339","type":"print"},{"value":"1531-2291","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]},"assertion":[{"value":"2026-02-16","name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}