Abstract:
Background. Colorectal cancer (CRC) remains a leading cause of cancer-related mortality
worldwide, and colonoscopy continues to represent the gold standard for CRC screening.
Recent advances, particularly the integration of artificial intelligence (AI) into clinical
practice, have enhanced screening efficiency.
Objective(s). A comparative analysis of the diagnostic efficacy and the potential to reduce
colorectal cancer morbidity using artificial intelligence-assisted versus conventional
colonoscopy.
Materials and methods. A narrative review was conducted of scientific publications
indexed in Google Scholar, PubMed, and Scopus, accessible up to 2025, focusing on the
application of artificial intelligence in colorectal cancer diagnosis. Relevant studies were
selected based on clinical applicability and the integration of AI technologies into clinical
practice.
Results. Studies have demonstrated that the use of artificial intelligence in lower
gastrointestinal endoscopy significantly improves diagnostic performance. The adenoma
detection rate (ADR) was approximately 1.5 times higher with advanced systems such as
EndoBRAIN (Olympus) and CAD EYE (Fujifilm) compared to conventional colonoscopy.
These findings have substantial implications for assessing endoscopic effectiveness. The
integration of CADe (Computer-Aided Detection) systems reduced adenoma miss rates
(AMR), with no significant differences between groups regarding the Boston Bowel
Preparation Scale or cecal intubation rate.
Conclusion(s). The findings further reinforce the role of artificial intelligence as a valuable
tool in modern colorectal endoscopy, underscoring its potential to significantly enhance
screening program efficiency, improve clinical outcomes and reduce colorectal cancerrelated mortality.
