Bowel preparation assessment using artificial intelligence: Systematic review.

Area of Science:

• Gastroenterology
• Medical Imaging
• Artificial Intelligence

Background:

• Insufficient bowel preparation is a primary cause of missed adenomas during colonoscopy.
• The Boston Bowel Preparation Scale (BBPS) is the standard for assessing bowel prep quality.
• Artificial intelligence (AI) offers potential for objective and automated bowel preparation assessment.

Purpose of the Study:

• To systematically review and evaluate Artificial Intelligence-based Bowel Preparation Scale (AI-BPS) systems for colonoscopy.
• To identify and describe existing AI-BPS technologies and their performance metrics.

Main Methods:

• A systematic literature review was conducted across MEDLINE, EMBASE, and SCOPUS.
• Search terms included colonoscopy, Bowel Preparation Scale (BPS), and Artificial Intelligence (AI).
• Two independent reviewers extracted data from eligible studies.

Main Results:

• Eight studies met the eligibility criteria, evaluating six AI-BPS systems.
• Most AI-BPS systems demonstrated comparable or superior performance to expert BBPS ratings.
• Three studies showed correlations between AI-BPS and adenoma/polyp detection rates.

Conclusions:

• AI-BPS holds promise for standardizing bowel preparation assessment and potentially outperforming human evaluations.
• Further research adhering to reporting guidelines is necessary for robust comparisons and meta-analysis.