Development and Validation of an Automated, Real-time Adenoma Detection Rate and Colonoscopy Quality Metrics...

Area of Science:

• Gastroenterology
• Medical Informatics
• Public Health

Background:

• High-quality colonoscopy is crucial for reducing colorectal cancer mortality.
• The adenoma detection rate (ADR) is a key quality indicator but is labor-intensive to calculate.
• Manual calculation of colonoscopy quality metrics is time-consuming and prone to error.

Purpose of the Study:

• To develop and validate a fully automated platform for calculating the adenoma detection rate (ADR).
• To assess the platform's ability to compute other essential colonoscopy quality indicators.
• To streamline colonoscopy quality assessment without manual data entry.

Main Methods:

• Collected endoscopy and pathology reports from six centers.
• Developed text-based classification parameters for identifying key colonoscopy metrics.
• Integrated automated calculators into a web-based platform and benchmarked against manual review.

Main Results:

• The automated platform achieved an ADR of 45.1%, closely matching manual review (44.3%).
• Demonstrated high predictive value for ADR-qualifying colonoscopies (sensitivity 0.918, specificity 1.0).
• Accurately reported cecal intubation rate (95.8%) and average withdrawal time (10:05 min).

Conclusions:

• A novel, fully automated platform for colonoscopy quality reporting is feasible and performs accurately.
• The platform integrates pathology report data for automated metric calculation.
• This system offers a scalable solution for improving colonoscopy quality across diverse practice settings.