Clinical Research Directory

The Efficacy of AI-Driven Feces Identification for Bowel Preparation Prior to Colonoscopy

Sponsor: Fu Jen Catholic University Hospital

Summary

The objective is to enhance diagnostic outcomes by ensuring thorough bowel cleanliness through AI-driven stool identifying system for bowel preparation in subjects undergoing colonoscopy.Colonoscopy is a key procedure for the prevention and early detection of colorectal cancer, with its diagnostic accuracy highly dependent on the quality of bowel preparation. Inadequate bowel cleansing can lead to missed lesions, prolonged procedure times, and the need for repeat examinations. Despite public health efforts that have improved screening rates over the past decade, the adequacy of bowel preparation has remained relatively unchanged, posing a persistent clinical challenge. With the rapid advancement of artificial intelligence (AI) technologies, deep learning-based image recognition has demonstrated outstanding performance in medical imaging applications. Recent studies have shown that AI can assist in real-time evaluation of stool appearance, providing patients with immediate feedback and personalized instructions to improve bowel preparation quality. Integrating AI systems into bowel cleansing protocols has the potential to enhance patient compliance, optimize bowel cleanliness, and consequently improve the diagnostic yield of colonoscopy. This study aims to evaluate the efficacy of an AI-based stool identifying system (AI-SIS), combined with the use of a Prepackaged Low Residue Diet (PLD) and standard bowel preparation instructions. Through a prospective, randomized, evaluator-blind, parallel-group clinical trial design, the study seeks to generate scientific evidence supporting the integration of AI technology into routine bowel preparation practices.

Official title: A Prospective, Randomized, Evaluator Blind, Parallel Study of the Efficacy of AI-driven Feces Identifying for the Bowel Preparation Prior to Colonoscopy

Key Details

Conditions

Interventions

Locations (1)