Evaluation of the clinical reasoning of GPT-4o, a multimodal generative artificial intelligence model, in 18 public gastroenterology case studies
clinical images did not increase the error rate; however, the observed failures revealed that the model tends to omit information already provided (such as clinical context or temporal criteria), thereby compromising the quality of its reasoning.
Authors
- Eloy Opisso (ORCID: https://orcid.org/0000-0002-6868-6737)
- N. Del Cacho (ORCID: https://orcid.org/0000-0002-9398-1826)
- Alejandro García‐Rudolph (ORCID: https://orcid.org/0000-0003-0853-8334)
- Elena Hernández-Pena (ORCID: https://orcid.org/0000-0002-1058-894X)
- Claudia Teixidó-Font
- Marc Navarro-Berenguel
Institutions
- Institut Guttmann (ES)
Publication Details
- Journal
- Revista Española de Enfermedades Digestivas
- Published
- 2025-01-01
- DOI
- https://doi.org/10.17235/reed.2025.11369/2025
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00