BACKGROUND
Artificial Intelligence may detect colorectal polyps that have been missed due to perceptual pitfalls. By reducing such miss rate, Artificial Intelligence may increase the detection of colorectal neoplasia leading to a higher degree of Colorectal Cancer (CRC) prevention.
METHODS
Patients undergoing CRC screening or surveillance were enrolled in 8 centres (Italy, UK, US), and randomized (1:1) to undergo two same-day, back-to-back colonoscopies with or without Artificial Intelligence (AI, deep learning CADe) in 2 different arms, namely AI followed by colonoscopy without AI or vice-versa. Adenoma miss rate (AMR) was calculated as the number of histologically-verified lesions detected at second colonoscopy divided by the total number of lesions detected at first and second colonoscopy. Mean number of lesions detected in the second colonoscopy and proportion of false negative subjects (no lesion at first colonoscopy and at least 1 at second) were calculated. Odds Ratio and 95% Confidence Intervals were adjusted by endoscopist, age, sex, and indication for colonoscopy. Adverse events were also measured.
RESULTS
A total of 230 subjects (116 AI first, 114 Std colonoscopy first) were included in the study analysis. AMR was 15.5% (38/246) and 32.4% (80/247) in the arm with AI and non-AI colonoscopy first, respectively (Adjusted OR [95% CI]: 0.38 [0.23; 0.62]). In detail, AMR was lower for AI-first for the <5 mm (15.9% vs 35.8%; OR: 0.34 [0.21; 0.55]) and non-polypoid lesions (16.8% vs. 45.8%; OR: 0.24[0.13; 0.43]), and it was lower both in the proximal (18.3% vs. 32.5%; OR: 0.46 [0.26; 0.78]) and distal colon (10.8% vs. 32.1%; OR: 0.25 [0.11; 0.57]). Mean number of adenomas at second colonoscopy was lower in the AI-first group as compared to non-AI colonoscopy first (0.33±0.63 vs 0.70±0.97, p<0.001). False negative rates were 6.8% (3/44 patients) and 29.6% (13/44) in the AI and non-AI first arms, respectively (OR [95% CI]: 0.17 [0.05; 0.67]). No difference in the rate of AE was found between the two groups.
CONCLUSIONS
AI resulted in an approximately two-fold reduction in miss rate of colorectal neoplasia, supporting AI-benefit in reducing perceptual errors for small and subtle lesions at standard colonoscopy.