Study of AI-Integrated Capsule Gastroscopy for Detecting Stomach Diseases at Home Compared to Standard Gastroscopy

What this Trial Is About

Researchers are evaluating the diagnostic accuracy of an AI-integrated Capsule Gastroscopy ACG system called GICE-1000 for detecting upper gastrointestinal UGI abnormalities in simulated home-use conditions. This study compares the ACG systems results to conventional esophagogastroduodenoscopy EGD, the current gold standard, to determine if the ACG can be a reliable and comfortable alternative for screening gastric diseases. It also assesses if AI-assisted reading of ACG videos improves diagnostic accuracy or reduces reading time compared to manual review. Participants will undergo the ACG examination in a simulated home environment after fasting and taking medications to clear gastric mucus. They follow video instructions to swallow the capsule and adjust body positions during the procedure. The capsule data is uploaded for analysis. Within 24 hours, participants also receive an EGD performed by experienced endoscopists blinded to ACG results. The ACG videos are reviewed both manually and with AI assistance to document findings and compare reading efficiency. During the study, participants provide informed consent, undergo fasting and preparation, and complete both ACG and EGD procedures. Researchers collect detailed imaging data, record abnormalities, and evaluate factors such as image quality and gastric cleanliness. The main outcome is the diagnostic accuracy of ACG for common UGI abnormalities. Secondary outcomes include comparison of diagnostic time between AI-assisted and manual readings and other measures related to examination completeness, image quality, and participant satisfaction. The total study duration includes screening, intervention, and follow-up periods, ending with data analysis and reporting.

Capsule Gastric Endoscopy for Gastric Disease Screening in Simulated Home Scenarios