[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-456604-105":59,"doc-detail-456604-en":135},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":128,"head_meta":130,"extra_data":132,"updated_unix":134},105,"en","diagnostic-capabilities-of-large-language-models-in-the-detection-of-scaphoid-fractures-in-the-emergency-department","Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department","","Scaphoid fractures represent 60%–70% of wrist traumas, and delayed diagnosis can lead to avascular necrosis and lasting functional impairment. Traditional radiographic interpretation is difficult due to anatomical complexity and overlapping structures. This retrospective study evaluated three next-generation large language models—ChatGPT-4o, Gemini 2.0, and Claude 3.5—for identifying scaphoid fractures and inferring surgical indications from wrist radiographs in 300 emergency patients. Claude 3.5 showed higher sensitivity but overall accuracy remained insufficient for independent clinical use.",{"@graph":69,"@context":127},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":35,"@type":76,"position":81},"https://docshare.wps.com/document/healthcare/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/diagnostic-capabilities-of-large-language-models-in-the-detection-of-scaphoid-fractures-in-the-emergency-department/456604/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/diagnostic-capabilities-of-large-language-models-in-the-detection-of-scaphoid-fractures-in-the-emergency-department/456604.png","ImageObject",300,407,{"name":92,"@type":93},"ahduag","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-09-30",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",11,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119,123],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the study address regarding scaphoid fractures in emergency care?","Question",{"text":113,"@type":114},"Scaphoid fractures are common in wrist trauma, but radiographic diagnosis is challenging, and delays can cause avascular necrosis and impaired function.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How were the large language models evaluated in this study?",{"text":118,"@type":114},"Three-view wrist radiographs were provided to each model on separate days, and diagnostic performance was assessed with overall, strict, and ideal accuracy, plus response consistency using Fleiss' kappa.",{"name":120,"@type":111,"acceptedAnswer":121},"What were the key findings about model performance for fracture detection?",{"text":122,"@type":114},"Claude 3.5 achieved the highest sensitivity (57.1%) compared with Gemini 2.0 (18.2%) and ChatGPT-4o (9.1%), while specificity stayed uniformly low across models.",{"name":124,"@type":111,"acceptedAnswer":125},"Can current LLMs be used independently for clinical decision-making for scaphoid fractures?",{"text":126,"@type":114},"No. The study concludes that diagnostic accuracy is insufficient for independent clinical use, though moderate performance in surgical decision-making suggests potential as assistive tools when combined with specialist expertise.","https://schema.org",{"og:url":83,"og:type":129,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":131,"canonical":83},"index,follow",{"doc_id":133,"site_id":62},456604,1790791357,{"code":4,"msg":5,"data":136},{"doc_id":133,"user_id":137,"nickname":92,"user_avatar":138,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":67,"doc_content":139,"file_id":140,"file_url":141,"file_type":142,"file_size":143,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":39,"language":144,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":145,"faqs":146,"seo_title":147,"seo_description":67,"update_tm":148,"read_time":46,"preview_page_count":4},3985747870414,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","ORIGINAL ARTICLE  \nDiagnostic capabilities of large language models in the detection ofscaphoid fractures in the emergency department  \n Bensu Bulut,1  Mehmet Yortanlı,2  Ayşenur Gür,3  Medine Akkan Öz,1  Hüseyin Mutlu4  \n1Department of Emergency, Gülhane Education and Research Hospital, Ankara-Türkiye 2Department of Emergency, Konya Numune Hospital, Konya-Türkiye  \n3Department of Emergency, Etimesgut Şehit Sait Ertürk State Hospital, Ankara-Türkiye 4Department of Emergency, Aksaray University Faculty of Medicine, Aksaray-Türkiye  \nABSTRACT  \nBACKGROUND: Scaphoid fractures account for 60%-70% of wrist traumas, with delayed diagnosis leading to avascular necrosis and functional impairment. Traditional radiographic assessment remains challenging due to anatomical complexity and overlapping structures. This study evaluated three next-generation large language models (LLMs) (ChatGPT-4o, Gemini 2.0, and Claude 3.5) for their ability to detect scaphoid fractures and determine surgical indications.  \nMETHODS: A retrospective observational study was conducted at Ankara Etlik City Hospital (October 2022 – January 2025) including 300 patients (150 with computed tomography confirmed (CT-confirmed) scaphoid fractures and 150 without fractures), aged 18-65 years, who presented to the emergency department (ED) with wrist trauma. Three-view wrist radiographs were presented to each LLM on three separate days. Diagnostic accuracy was assessed using overall accuracy (all three responses correct), strict accuracy (≥2 correct responses), and ideal accuracy (≥1 correct response). Response consistency was evaluated using Fleiss' kappa coefficient. Surgical indications were determined based on fracture displacement criteria.  \nRESULTS: Claude 3.5 demonstrated superior sensitivity (57.1%) compared to Gemini 2.0 (18.2%) and ChatGPT-4o (9.1%) for fracture detection (p\u003C0.001). Ideal accuracy rates were 79.3%, 36.0%, and 17.3%, respectively. Specificity remained uniformly low across models (43.1%-43.8%). All models performed better in non-fracture cases, with ideal accuracy exceeding 83%. Response consistency was moderate for all models (κ=0.36-0.41). For surgical indication assessment, Claude 3.5 identified 37.0% of cases requiring surgery, compared to ChatGPT-4o (34.1%) and Gemini 2.0 (24.4%), with correct determination rates of 73.7%, 71.4%, and 80.0%, respectively. CONCLUSION: Current LLMs demonstrate insufficient diagnostic accuracy for independent clinical use in scaphoid fracture detection. Claude 3.5's 57.1% sensitivity indicates that these technologies require substantial improvement before clinical deployment. However, their moderate performance in surgical decision-making suggests potential utility as assistive tools when combined with specialist expertise. Further development focusing on musculoskeletal-specific training is essential.  \nKeywords: Artificial intelligence; diagnostic accuracy; large language models; scaphoid fractures; wrist radiography.  \nINTRODUCTION  \nScaphoid fractures account for approximately 60%-70% of all wrist traumas presenting to the emergency department (ED)  \nand predominantly affect the male population.[1,2] These fractures typically occur following a fall on an outstretched hand or sudden forced dorsiflexion of the wrist.[3,4] Timely and ac-  \nCite this article as: Bulut B, Yortanlı M, GürA, Öz MA, Mutlu H. Diagnostic capabilities of large language models in the detection ofscaphoid fractures in the emergency department. Ulus Travma Acil Cerrahi Derg 2025;31:987-994.  \nAddress for correspondence: Ayşenur Gür  \nDepartment of Emergency, Etimesgut Şehit Sait Ertürk State Hospital, Ankara, Türkiye  \nE-mail: [draysenurcakici@gmail.com](draysenurcakici@gmail.com)  \nUlus Travma Acil Cerrahi Derg 2025;31(10):987-994 DOI: 10.14744/tjtes.2025.98680  \nSubmitted: 09.09.2025 Revised: 22.09.2025 Accepted: 23.09.2025 Published: 07.10.2025  \nOPEN ACCESS This is an open access article under the CC BY-NC license ([http://creativec","cbCaiePj5gjLsnJ0","https://ap.wps.com/l/cbCaiePj5gjLsnJ0","pdf",626004,"English","# Abstract\n# Background\n# Methods\n# Results\n# Conclusion\n# Introduction","[{\"question\":\"What problem does the study address regarding scaphoid fractures in emergency care?\",\"answer\":\"Scaphoid fractures are common in wrist trauma, but radiographic diagnosis is challenging, and delays can cause avascular necrosis and impaired function.\"},{\"question\":\"How were the large language models evaluated in this study?\",\"answer\":\"Three-view wrist radiographs were provided to each model on separate days, and diagnostic performance was assessed with overall, strict, and ideal accuracy, plus response consistency using Fleiss' kappa.\"},{\"question\":\"What were the key findings about model performance for fracture detection?\",\"answer\":\"Claude 3.5 achieved the highest sensitivity (57.1%) compared with Gemini 2.0 (18.2%) and ChatGPT-4o (9.1%), while specificity stayed uniformly low across models.\"},{\"question\":\"Can current LLMs be used independently for clinical decision-making for scaphoid fractures?\",\"answer\":\"No. The study concludes that diagnostic accuracy is insufficient for independent clinical use, though moderate performance in surgical decision-making suggests potential as assistive tools when combined with specialist expertise.\"}]","Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department | PDF",1790746489]