EUVIMEDEUVIMED entdecken
BeobachtungsstudieOpen Access

Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department

Bensu Bulut, Mehmet Yortanlı, Ayşenur Gür, Medine Akkan Öz, Hüseyin Mutlu

Turkish Journal of Trauma and Emergency Surgery · 2025-10 · 31 (10) · 987-994

DOI: 10.14744/tjtes.2025.98680 ↗

Kurz gefasst

Beobachtungsstudie in der Zeitschrift Turkish Journal of Trauma and Emergency Surgery, 2025-10. Verfasst von Bensu Bulut und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Emergency, Gülhane Education and Research Hospital, Ankara-Türkiye. Hauptthemen laut MEDLINE: Scaphoid Bone, Fractures, Bone, Wrist Injuries. Die Arbeit wurde bisher 6-mal zitiert (Crossref). Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: 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. METHODS: 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. RESULTS: Claude 3.5 demonstrated superior sensitivity (57.1%) compared to Gemini 2.0 (18.2%) and ChatGPT-4o (9.1%) for fracture detection (p

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Scaphoid BoneFractures, BoneWrist InjuriesHumansAdultEmergency Service, HospitalRetrospective StudiesMiddle AgedMaleFemaleAdolescentYoung AdultAgedTomography, X-Ray ComputedSensitivity and SpecificityLarge Language Models

Autor:innen und Institutionen

  • Department of Emergency, Gülhane Education and Research Hospital, Ankara-Türkiye.
  • Department of Emergency, Konya Numune Hospital, Konya-Türkiye.
  • Department of Emergency, Etimesgut Şehit Sait Ertürk State Hospital, Ankara-Türkiye.
  • Department of Emergency, Aksaray University Faculty of Medicine, Aksaray-Türkiye.

Zitieren

APA

Bulut, B., Yortanlı, M., Gür, A., Akkan Öz, M., & Mutlu, H. (2025). Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department. Turkish Journal of Trauma and Emergency Surgery, 31(10), 987-994. https://doi.org/10.14744/tjtes.2025.98680

Vancouver

Bulut B, Yortanlı M, Gür A, Akkan Öz M, Mutlu H. Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department. Turkish Journal of Trauma and Emergency Surgery. 2025;31(10):987-994. doi:10.14744/tjtes.2025.98680

BibTeX
@article{Bulut2025,
  author = {Bulut, Bensu and Yortanlı, Mehmet and Gür, Ayşenur and Akkan Öz, Medine and Mutlu, Hüseyin},
  title = {Diagnostic capabilities of large language models in the detection of scaphoid fractures in the emergency department},
  journal = {Turkish Journal of Trauma and Emergency Surgery},
  year = {2025},
  volume = {31},
  number = {10},
  pages = {987--994},
  doi = {10.14744/tjtes.2025.98680}
}
Kostenlos starten →