Michael B Wallace1, Prateek Sharma2, Pradeep Bhandari3, James East4, Giulio Antonelli5, Roberto Lorenzetti6, Micheal Vieth7, Ilaria Speranza8, Marco Spadaccini6, Madhav Desai3, Frank J Lukens9, Genci Babameto10, Daisy Batista10, Davinder Singh10, William Palmer9, Francisco Ramirez11, Rebecca Palmer4, Tisha Lunsford11, Kevin Ruff11, Elizabeth Bird-Liebermann4, Victor Ciofoaia10, Sophie Arndtz3, David Cangemi9, Kirsty Puddick3, Gregory Derfus12, Amitpal S Johal13, Mohammed Barawi14, Luigi Longo15, Luigi Moro15, Alessandro Repici16, Cesare Hassan16. 1. Division of Gastroenterology and Hepatology, Mayo Clinic Jacksonville, Florida; Division of Gastroenterology, Sheikh Shakhbout Medical City (SSMC), Abu Dhabi, UAE. Electronic address: mwallace@ssmc.ae. 2. Department of Gastroenterology and Hepatology, University of Kansas Medical Center, Kansas City, Kansas. 3. Division of Gastroenterology, Queen Alexandra Hospital, Portsmouth, UK. 4. Translational Gastroenterology Unit, John Radcliffe Hospital, Oxford, UK. 5. Gastroenterology Unit, Nuovo Regina Margherita Hospital, Rome, Italy; Department of Anatomical, Histological, Forensic Medicine and Orthopedics Sciences, "Sapienza" University of Rome, Italy; Gastroenterology and Digestive Endoscopy Unit, Ospedale dei Castelli Hospital, Ariccia, Rome, Italy. 6. Gastroenterology Unit, Nuovo Regina Margherita Hospital, Rome, Italy. 7. Institut für Pathologie Klinikum Bayreuth GmbH, Bayreuth, Germany. 8. Cros NT, Verona, Italy. 9. Division of Gastroenterology and Hepatology, Mayo Clinic Jacksonville, Florida. 10. Division of Gastroenterology and Hepatology, Mayo Clinic LaCrosse, LaCrosse, Wisconsin. 11. Division of Gastroenterology and Hepatology, Mayo Clinic Scottsdale, Scottsdale, Arizona. 12. Division of Gastroenterology and Hepatology, Mayo Clinic Eau Claire, Eau Claire, Wisconsin. 13. Division of Gastroenterology, Geisinger Medical Center, Danville, Pennsylvania. 14. Gastroenterology & Digestive Health, Ascension St. John Hospital, Detroit, Michigan. 15. Cosmo Artificial Intelligence-AI Ltd, Dublin, Ireland. 16. Department of Biomedical Sciences, Humanitas University, Rozzano, Milan, Italy; Endoscopy Unit, Humanitas Clinical and Research Center IRCCS, Rozzano, Milan, Italy.
Abstract
BACKGROUND & AIMS: Artificial intelligence (AI) may detect colorectal polyps that have been missed due to perceptual pitfalls. By reducing such miss rate, AI 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 centers (Italy, UK, US), and randomized (1:1) to undergo 2 same-day, back-to-back colonoscopies with or without AI (deep learning computer aided diagnosis endoscopy) 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 ratios (ORs) and 95% confidence intervals (CIs) 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 standard colonoscopy first) were included in the study analysis. AMR was 15.5% (38 of 246) and 32.4% (80 of 247) in the arm with AI and non-AI colonoscopy first, respectively (adjusted OR, 0.38; 95% CI, 0.23-0.62). In detail, AMR was lower for AI first for the ≤5 mm (15.9% vs 35.8%; OR, 0.34; 95% CI, 0.21-0.55) and nonpolypoid lesions (16.8% vs 45.8%; OR, 0.24; 95% CI, 0.13-0.43), and it was lower both in the proximal (18.3% vs 32.5%; OR, 0.46; 95% CI, 0.26-0.78) and distal colon (10.8% vs 32.1%; OR, 0.25; 95% CI, 0.11-0.57). Mean number of adenomas at second colonoscopy was lower in the AI-first group as compared with non-AI colonoscopy first (0.33 ± 0.63 vs 0.70 ± 0.97, P < .001). False negative rates were 6.8% (3 of 44 patients) and 29.6% (13 of 44) in the AI and non-AI first arms, respectively (OR, 0.17; 95% CI, 0.05-0.67). No difference in the rate of adverse events was found between the 2 groups. CONCLUSIONS: AI resulted in an approximately 2-fold reduction in miss rate of colorectal neoplasia, supporting AI-benefit in reducing perceptual errors for small and subtle lesions at standard colonoscopy. CLINICALTRIALS: gov, Number: NCT03954548.
BACKGROUND & AIMS: Artificial intelligence (AI) may detect colorectal polyps that have been missed due to perceptual pitfalls. By reducing such miss rate, AI 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 centers (Italy, UK, US), and randomized (1:1) to undergo 2 same-day, back-to-back colonoscopies with or without AI (deep learning computer aided diagnosis endoscopy) 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 ratios (ORs) and 95% confidence intervals (CIs) 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 standard colonoscopy first) were included in the study analysis. AMR was 15.5% (38 of 246) and 32.4% (80 of 247) in the arm with AI and non-AI colonoscopy first, respectively (adjusted OR, 0.38; 95% CI, 0.23-0.62). In detail, AMR was lower for AI first for the ≤5 mm (15.9% vs 35.8%; OR, 0.34; 95% CI, 0.21-0.55) and nonpolypoid lesions (16.8% vs 45.8%; OR, 0.24; 95% CI, 0.13-0.43), and it was lower both in the proximal (18.3% vs 32.5%; OR, 0.46; 95% CI, 0.26-0.78) and distal colon (10.8% vs 32.1%; OR, 0.25; 95% CI, 0.11-0.57). Mean number of adenomas at second colonoscopy was lower in the AI-first group as compared with non-AI colonoscopy first (0.33 ± 0.63 vs 0.70 ± 0.97, P < .001). False negative rates were 6.8% (3 of 44 patients) and 29.6% (13 of 44) in the AI and non-AI first arms, respectively (OR, 0.17; 95% CI, 0.05-0.67). No difference in the rate of adverse events was found between the 2 groups. CONCLUSIONS: AI resulted in an approximately 2-fold reduction in miss rate of colorectal neoplasia, supporting AI-benefit in reducing perceptual errors for small and subtle lesions at standard colonoscopy. CLINICALTRIALS: gov, Number: NCT03954548.