The study aims to evaluate the effectiveness of artificial intelligence-assisted colonoscopy in increasing adenoma detection rate and the accuracy in the characterization of colorectal lesions, compared to standard colonoscopy, in a randomized controlled clinical trial setting.
All patients aged 18 years or older, with an elective indication for colonoscopy who sign the informed consent form agreeing to participate in the study.
History of inflammatory bowel disease.
History of colorectal cancer.
Personal history of colorectal surgery.
Contraindication to endoscopic biopsies.
Parallel
Colonoscopy
Procedure/SurgeryThis single-center, randomized, open-label clinical trial will assess the effectiveness of artificial intelligence (AI)-assisted colonoscopy versus standard high-definition colonoscopy in detecting and characterizing colorectal lesions. Conducted over 12 months in São Paulo, Brazil, the study will include 100 adult patients undergoing elective colonoscopy. Participants will be stratified by age and randomized (1:1) after sedation. All lesions will be resected, recorded, and analyzed histologically. The intervention group will also include AI output data (CAD EYE). The primary goals are to evaluate adenoma detection rate (ADR) and AI diagnostic accuracy. Given the global burden of colorectal cancer (CRC), particularly in developing countries, this study aims to provide real-world data on the impact of AI in CRC screening.
Number of patients with at least one adenoma detected, confirmed by histopathological analysis, during colonoscopy, in the AI group vs. control group
The measure will be expressed as the number and percentage (%) of patients with at least one adenoma detected during colonoscopy and confirmed by histopathological analysis, comparing the AI and non-AI groups (CAD EYE). Detection will be based on the analysis of biopsies performed and processed according to the standard protocol.
Diagnostic accuracy of CAD EYE for characterization of lesions as neoplastic (adenoma) or non-neoplastic (hyperplastic), compared to histopathological analysis as the gold standard.
The accuracy of artificial intelligence (CAD EYE) in characterizing detected lesions as neoplastic or non-neoplastic will be calculated, based on comparison with histopathological diagnosis (gold standard). The following will be reported: sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), in percentage (%), for each type of lesion.
Click on the lead sponsor to view all of their trials.
