Actively Recruiting
Randomized Trial of AI-Assisted Versus Conventional EGGIM Scoring Training for Endoscopists Assessing Gastric Intestinal Metaplasia
Led by Qilu Hospital of Shandong University · Updated on 2026-09-08
8
Participants Needed
1
Research Sites
13 weeks
Total Duration
AI-Summary
What This Trial Is About
This trial studies the condition of gastric intestinal metaplasia and evaluates the effectiveness of an AI system designed to train endoscopists in assessing this condition using the EGGIM score. The research is a prospective randomized controlled trial with a crossover design comparing AI-assisted scoring with conventional scoring methods. The study aims to improve the diagnostic accuracy of endoscopists with different experience levels by analyzing key performance metrics such as AUC, sensitivity, and specificity. Participants are endoscopists who will review image-enhanced endoscopy videos of five standardized gastric areas. They are randomly assigned to two groups one group uses the AI-assisted EGGIM scoring first and then conventional scoring after a washout period, while the other group does the reverse. The interventions involve diagnostic tests where videos are evaluated either with or without AI assistance. This crossover design allows comparison of scoring performance within the same participants over time. During the trial, endoscopists scoring performances will be assessed through their evaluations of the videos in both AI-assisted and conventional phases. The primary outcome focuses on the improvement in EGGIM scoring ability with AI assistance over approximately three months. Secondary outcomes include performance differences based on endoscopists experience levels. Monitoring of diagnostic accuracy and related metrics will be conducted throughout the study duration.
CONDITIONS
Brief Title
Efficacy of an AI System in Training Endoscopists to Assess Gastric Intestinal Metaplasia Via the EGGIM Score
Research Team
Z
Zhen Li
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here