Bot vs. doc—who is better at reading proximal humerus fracture x-rays?
This study demonstrates that ChatGPT-5 is highly inaccurate at identifying PHFx on shoulder x-rays, characterizing the fracture patterns, and providing accurate interpretations of shoulder x-rays. Over-reliance on generative artificial intelligence to guide clinical decisions risks harm to the patients and should be approached with limited credence.
Authors
- Andrew J. Luzzi (ORCID: https://orcid.org/0000-0001-9714-3873)
- Hasani W. Swindell (ORCID: https://orcid.org/0000-0003-3230-460X)
- Mark Ren (ORCID: https://orcid.org/0000-0003-0179-3471)
- Matthew R. LeVasseur (ORCID: https://orcid.org/0000-0003-3366-3639)
- William N. Levine (ORCID: https://orcid.org/0000-0002-2826-1179)
- Kyle K. Obana (ORCID: https://orcid.org/0000-0002-3965-8999)
Institutions
- Columbia University Irving Medical Center (US)
Publication Details
- Journal
- JSES International
- Published
- 2025-12-02
- DOI
- https://doi.org/10.1016/j.jseint.2025.101426
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00