{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T04:02:49Z","timestamp":1785556969851,"version":"3.56.0"},"reference-count":24,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100004375","name":"Tel Aviv University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100004375","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Arch Orthop Trauma Surg"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Cut-out is the most consequential mechanical complication after proximal femoral nailing and requires prompt recognition. General-purpose artificial intelligence (AI) models with image-interpretation capability are increasingly accessible, yet their diagnostic performance for this task is unknown. We evaluated ChatGPT\u2019s accuracy for predicting cut-out following proximal femoral nailing and compared TFNA and Gamma nail subgroups.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>In this retrospective predictive-accuracy study, 989 patients (683 TFNA, 306 Gamma nail; mean age 83.8 years; cut-out prevalence 2.5%) were analysed. For each case, three baseline radiographs \u2014 the injury anteroposterior and lateral views and the immediate postoperative control radiograph, all obtained before any complication could be radiographically present \u2014 together with the full clinical data set (excluding complication-related information) were submitted to ChatGPT, which returned a binary prediction (yes\/no) and probability estimate (0\u2013100%). The follow-up radiographs on which cut-out becomes apparent were not provided to the model. Ground truth was established on subsequent follow-up radiographs by a departmental review board requiring agreement of two senior surgeons. Accuracy metrics were calculated with Wilson 95% confidence intervals (CI); exact McNemar\u2019s and Mann-Whitney U tests assessed directional bias and calibration. Reporting followed STARD guidelines.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      ChatGPT showed a sensitivity of 68.0% (95% CI 48.4\u201382.8%), specificity of 62.7% (59.6\u201365.7%), PPV of 4.5% (2.8\u20137.1%), NPV of 98.7% (97.4\u201399.3%), and AUC of 0.694 (0.580\u20130.790). Performance was higher for TFNA (sensitivity 73.7%, AUC 0.730) than for Gamma nail (sensitivity 50.0%, AUC 0.606). The model over-predicted cut-out, with 360 false positives against 8 false negatives (45:1; McNemar\u2019s\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001, Holm-corrected). Calibration was inverted: the median predicted probability was lower for correct (9%) than for incorrect (40%) classifications (Mann-Whitney\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>ChatGPT demonstrated limited, clinically insufficient accuracy for predicting cut-out following proximal femoral nailing, with a prohibitive false-positive burden and inverted calibration. Subgroup point estimates were lower for Gamma nail cases, but this exploratory difference did not reach statistical significance. General-purpose AI models are not currently suitable as a substitute for, or adjunct to, surgeon-led surveillance of this complication. Task-specific training, external validation, and defined clinical boundaries are required before such tools can be considered for fracture follow-up pathways.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00402-026-06447-5","type":"journal-article","created":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T03:22:55Z","timestamp":1785554575000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Predictive accuracy of a general-purpose artificial intelligence model for cut-out following proximal femoral nailing"],"prefix":"10.1007","volume":"146","author":[{"given":"Guy","family":"Ben Arie","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yaniv","family":"Warschawski","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nissan","family":"Amzallag","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hadar","family":"Gan-Or","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tomer","family":"Ben-Tov","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amal","family":"Khoury","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Idit","family":"Horev","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nadav","family":"Graif","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guy","family":"Morag","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,8,1]]},"reference":[{"key":"6447_CR1","doi-asserted-by":"publisher","first-page":"1726","DOI":"10.1007\/s00198-006-0172-4","volume":"17","author":"O Johnell","year":"2006","unstructured":"Johnell O, Kanis JA (2006) An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int 17:1726\u20131733. https:\/\/doi.org\/10.1007\/s00198-006-0172-4","journal-title":"Osteoporos Int"},{"key":"6447_CR2","doi-asserted-by":"publisher","first-page":"1573","DOI":"10.1001\/jama.2009.1567","volume":"302","author":"CA Brauer","year":"2009","unstructured":"Brauer CA, Coca-Perraillon M, Cutler DM, Rosen AB (2009) Incidence and mortality of hip fractures in the United States. JAMA 302:1573\u20131579. https:\/\/doi.org\/10.1001\/jama.2009.1567","journal-title":"JAMA"},{"key":"6447_CR3","doi-asserted-by":"publisher","first-page":"2053","DOI":"10.1056\/NEJMcp1611090","volume":"377","author":"M Bhandari","year":"2017","unstructured":"Bhandari M, Swiontkowski M (2017) Management of acute hip fracture. N Engl J Med 377:2053\u20132062. https:\/\/doi.org\/10.1056\/NEJMcp1611090","journal-title":"N Engl J Med"},{"key":"6447_CR4","doi-asserted-by":"publisher","first-page":"1058","DOI":"10.2106\/00004623-199507000-00012","volume":"77","author":"MR Baumgaertner","year":"1995","unstructured":"Baumgaertner MR, Curtin SL, Lindskog DM, Keggi JM (1995) The value of the tip-apex distance in predicting failure of fixation of peritrochanteric fractures of the hip. J Bone Jt Surg Am 77:1058\u20131064","journal-title":"J Bone Joint Surg Am"},{"key":"6447_CR5","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1038\/s41598-021-04252-1","volume":"12","author":"G Caruso","year":"2022","unstructured":"Caruso G, Corradi N, Caldaria A et al (2022) New tip-apex distance and calcar-referenced tip-apex distance cut-offs may be the best predictors for cut-out risk after intramedullary fixation of proximal femur fractures. Sci Rep 12:357. https:\/\/doi.org\/10.1038\/s41598-021-04252-1","journal-title":"Sci Rep"},{"key":"6447_CR6","doi-asserted-by":"publisher","first-page":"1266","DOI":"10.2106\/JBJS.K.00357","volume":"94","author":"K De Bruijn","year":"2012","unstructured":"De Bruijn K, den Hartog D, Tuinebreijer W, Roukema G (2012) Reliability of predictors for screw cutout in intertrochanteric hip fractures. J Bone Joint Surg Am 94:1266\u20131272. https:\/\/doi.org\/10.2106\/JBJS.K.00357","journal-title":"J Bone Joint Surg Am"},{"key":"6447_CR7","doi-asserted-by":"publisher","first-page":"643","DOI":"10.2106\/00004623-200105000-00001","volume":"83","author":"GJ Haidukewych","year":"2001","unstructured":"Haidukewych GJ, Israel TA, Berry DJ (2001) Reverse obliquity fractures of the intertrochanteric region of the femur. J Bone Jt Surg Am 83:643\u2013650","journal-title":"J Bone Joint Surg Am"},{"key":"6447_CR8","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1097\/BOT.0000000000000961","volume":"31","author":"DJ Ciufo","year":"2017","unstructured":"Ciufo DJ, Zaruta DA, Lipof JS et al (2017) Risk factors associated with cephalomedullary nail cutout in the treatment of trochanteric hip fractures. J Orthop Trauma 31:583\u2013588. https:\/\/doi.org\/10.1097\/BOT.0000000000000961","journal-title":"J Orthop Trauma"},{"key":"6447_CR9","doi-asserted-by":"publisher","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","volume":"29","author":"AJ Thirunavukarasu","year":"2023","unstructured":"Thirunavukarasu AJ, Ting DSJ, Elangovan K et al (2023) Large language models in medicine. Nat Med 29:1930\u20131940. https:\/\/doi.org\/10.1038\/s41591-023-02448-8","journal-title":"Nat Med"},{"key":"6447_CR10","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1001\/jama.2023.8288","volume":"330","author":"Z Kanjee","year":"2023","unstructured":"Kanjee Z, Crowe B, Rodman A (2023) Accuracy of a generative artificial intelligence model in a complex diagnostic challenge. JAMA 330:78\u201380. https:\/\/doi.org\/10.1001\/jama.2023.8288","journal-title":"JAMA"},{"key":"6447_CR11","doi-asserted-by":"publisher","first-page":"e233391","DOI":"10.1001\/jamanetworkopen.2023.3391","volume":"6","author":"JR Lex","year":"2023","unstructured":"Lex JR, Di Michele J, Koucheki R, Pincus D, Whyne C, Ravi B (2023) Artificial intelligence for hip fracture detection and outcome prediction: a systematic review and meta-analysis. JAMA Netw Open 6:e233391","journal-title":"JAMA Netw Open"},{"key":"6447_CR12","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1148\/radiol.211785","volume":"304","author":"RYL Kuo","year":"2022","unstructured":"Kuo RYL, Harrison C, Curran TA, Jones B, Freethy A, Cussons D et al (2022) Artificial intelligence in fracture detection: a systematic review and meta-analysis. Radiology 304:50\u201362","journal-title":"Radiology"},{"key":"6447_CR13","doi-asserted-by":"publisher","first-page":"23053","DOI":"10.1038\/s41598-024-73058-8","volume":"14","author":"J Husarek","year":"2024","unstructured":"Husarek J, Hess S, Razaeian S, Ruder TD, Sehmisch S, M\u00fcller M et al (2024) Artificial intelligence in commercial fracture detection products: a systematic review and meta-analysis of diagnostic test accuracy. Sci Rep 14:23053","journal-title":"Sci Rep"},{"key":"6447_CR14","doi-asserted-by":"publisher","first-page":"1508","DOI":"10.1002\/jor.26109","volume":"43","author":"A Misir","year":"2025","unstructured":"Misir A, Yuce A (2025) Artificial intelligence in orthopaedic research: a comprehensive review. J Orthop Res 43:1508\u20131527. https:\/\/doi.org\/10.1002\/jor.26109","journal-title":"J Orthop Res"},{"key":"6447_CR15","doi-asserted-by":"publisher","first-page":"112570","DOI":"10.1016\/j.injury.2025.112570","volume":"56","author":"A Misir","year":"2025","unstructured":"Misir A (2025) Artificial intelligence in orthopaedic trauma: a comprehensive review. Injury 56:112570. https:\/\/doi.org\/10.1016\/j.injury.2025.112570","journal-title":"Injury"},{"key":"6447_CR16","doi-asserted-by":"publisher","first-page":"e0260471","DOI":"10.1371\/journal.pone.0260471","volume":"16","author":"SJ Federer","year":"2021","unstructured":"Federer SJ, Jones GG (2021) Artificial intelligence in orthopaedics: a scoping review. PLoS ONE 16:e0260471. https:\/\/doi.org\/10.1371\/journal.pone.0260471","journal-title":"PLoS ONE"},{"key":"6447_CR17","doi-asserted-by":"publisher","first-page":"2761","DOI":"10.1016\/j.arthro.2022.04.014","volume":"38","author":"PN Ramkumar","year":"2022","unstructured":"Ramkumar PN, Pang M, Polisetty T, Helm JM, Karnuta JM (2022) Meaningless applications and misguided methodologies in artificial intelligence-related orthopaedic research propagates hype over hope. Arthroscopy 38:2761\u20132766. https:\/\/doi.org\/10.1016\/j.arthro.2022.04.014","journal-title":"Arthroscopy"},{"key":"6447_CR18","doi-asserted-by":"publisher","unstructured":"Aghili N, Hughes K, Nazarpour K, Simpson H, Yang I (2025) The need for better quality studies: a systematic scoping review of current utility of artificial intelligence in orthopaedics and research gaps in the knee joint. https:\/\/doi.org\/10.1016\/j.knee.2025.10.022. Knee","DOI":"10.1016\/j.knee.2025.10.022"},{"issue":"Suppl 1","key":"6447_CR19","doi-asserted-by":"publisher","first-page":"S1","DOI":"10.1097\/BOT.0000000000001063","volume":"32","author":"EG Meinberg","year":"2018","unstructured":"Meinberg EG, Agel J, Roberts CS, Karam MD, Kellam JF (2018) Fracture and dislocation classification compendium \u2014 2018. J Orthop Trauma 32(Suppl 1):S1\u2013S170. https:\/\/doi.org\/10.1097\/BOT.0000000000001063","journal-title":"J Orthop Trauma"},{"key":"6447_CR20","doi-asserted-by":"publisher","DOI":"10.1097\/BOT.0000000000003166","author":"JA Holmstrom","year":"2026","unstructured":"Holmstrom JA, Braithwaite CL, Alhankawi AR et al (2026) Comparing the efficacy between ChatGPT 5, Grok 3, and Claude 4.5 Sonnet in analyzing orthopedic trauma-related imaging. J Orthop Trauma. https:\/\/doi.org\/10.1097\/BOT.0000000000003166","journal-title":"J Orthop Trauma"},{"key":"6447_CR21","doi-asserted-by":"publisher","first-page":"h5527","DOI":"10.1136\/bmj.h5527","volume":"351","author":"PM Bossuyt","year":"2015","unstructured":"Bossuyt PM, Reitsma JB, Bruns DE et al (2015) STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ 351:h5527. https:\/\/doi.org\/10.1136\/bmj.h5527","journal-title":"BMJ"},{"key":"6447_CR22","doi-asserted-by":"publisher","first-page":"2269","DOI":"10.1002\/ksa.12657","volume":"33","author":"FC Oettl","year":"2025","unstructured":"Oettl FC, Zsidai B, Oeding JF et al (2025) Beyond traditional orthopaedic data analysis: AI, multimodal models and continuous monitoring. Knee Surg Sports Traumatol Arthrosc 33:2269\u20132275. https:\/\/doi.org\/10.1002\/ksa.12657","journal-title":"Knee Surg Sports Traumatol Arthrosc"},{"key":"6447_CR23","doi-asserted-by":"publisher","first-page":"2090","DOI":"10.1016\/j.arthro.2021.12.030","volume":"38","author":"KN Kunze","year":"2022","unstructured":"Kunze KN, Krivicich LM, Clapp IM et al (2022) Machine learning algorithms predict achievement of clinically significant outcomes after orthopaedic surgery: a systematic review. Arthroscopy 38:2090\u20132105. https:\/\/doi.org\/10.1016\/j.arthro.2021.12.030","journal-title":"Arthroscopy"},{"key":"6447_CR24","doi-asserted-by":"publisher","unstructured":"Patel AA, Schwab JH, Amanatullah DF, Divi SN (2023) AOA Critical Issues Symposium: shaping the impact of artificial intelligence within orthopaedic surgery. J Bone Jt Surg Am 105:1475\u20131479. https:\/\/doi.org\/10.2106\/JBJS.22.01330","DOI":"10.2106\/JBJS.22.01330"}],"container-title":["Archives of Orthopaedic and Trauma Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00402-026-06447-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00402-026-06447-5","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00402-026-06447-5.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T03:22:59Z","timestamp":1785554579000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00402-026-06447-5"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":24,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["6447"],"URL":"https:\/\/doi.org\/10.1007\/s00402-026-06447-5","relation":{},"ISSN":["1434-3916"],"issn-type":[{"value":"1434-3916","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,1]]},"assertion":[{"value":"17 June 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 July 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 August 2026","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","label":"Conflict of interest","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"276"}}