What question did this study set out to answer?

To evaluate the performance of GastroGPT, a specialized AI for gastroenterology, in clinical tasks.

November 1, 2025JGH Open0 citationsOpen Access

GastroGPT: Exploring AI's Role and Challenges in Gastroenterology

GastroGPT Pioneering Specialized AI in Gastroenterology: Strengths, Pitfalls, and the Road to Clinical Integration

View Full Paper

Ask AI

Bookmark

Authors

GastroGPT demonstrates improved clinical task performance in inflammatory bowel disease cases, suggesting potential for enhanced patient care.

To evaluate the performance of GastroGPT, a specialized AI for gastroenterology, in clinical tasks.
Developed a transformer-based large language model specifically for gastroenterology.
Tested GastroGPT on simulated cases involving inflammatory bowel disease and endoscopy.
Utilized a dataset of 1.2 million tokens from peer-reviewed content and clinical guidelines.
GastroGPT achieved a mean score of 8.1 ± 1.8 on a 10-point Likert scale across various tasks.
It outperformed general-purpose models like GPT-4, Bard, and Claude in clinical tasks (p < 0.001).
Demonstrated higher scores in tasks such as patient history acquisition and patient education.

Tiwari et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

November 1, 2025JGH Open0 citationsOpen Access

GastroGPT Pioneering Specialized AI in Gastroenterology: Strengths, Pitfalls, and the Road to Clinical Integration

View Full Paper

Ask AI

Bookmark

Authors

GastroGPT demonstrates improved clinical task performance in inflammatory bowel disease cases, suggesting potential for enhanced patient care.

To evaluate the performance of GastroGPT, a specialized AI for gastroenterology, in clinical tasks.
Developed a transformer-based large language model specifically for gastroenterology.
Tested GastroGPT on simulated cases involving inflammatory bowel disease and endoscopy.
Utilized a dataset of 1.2 million tokens from peer-reviewed content and clinical guidelines.
GastroGPT achieved a mean score of 8.1 ± 1.8 on a 10-point Likert scale across various tasks.
It outperformed general-purpose models like GPT-4, Bard, and Claude in clinical tasks (p < 0.001).
Demonstrated higher scores in tasks such as patient history acquisition and patient education.

Tiwari et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark