S578 Comparison of Effectiveness of Artificial Intelligence (AI)-Aided Colonoscopy With Conventional Colonoscopy in Enhanced Detection of Colorectal Adenomas and Polyps—A Systematic Review and Meta-Analysis

Introduction: Artificial intelligence (AI)-aided colonscopy has shown significant improvement in detection of precancerous lesions. This systematic review and meta-analysis evaluates its performance in comparison to regular conventional colonoscopy by assessing the adenoma and polyp detection rates, as well as adenomas and polyps detected per procedure, thus highlighting the potential of AI to enhance prevention of colorectal cancer. Methods: The review conducted follows the PRISMA guidelines and major medical databases, which include PUBMED, Google Scholar and Science-Direct, were extensively searched using a comprehensive search term to identify and retrieve available articles. The articles that included the assessment of the detection of adenoma and polyp by AI-assisted colonoscopy and traditional colonoscopy were included in the final analysis. The data was analyzed using the Meta, Metadata and the Metafor packages of R Studio. The odds ratio (OR) of detection of adenoma and polyp through different colonoscopy techniques was assessed as the primary outcome. The Mantel-Haenszel method and the Inverse variance method were utilized to analyze the odds ratio. The I2 test was used to assess the heterogeneity of the studies. Results: The study included a total of 16 studies with 5,466 patients undergoing AI-assisted colonoscopy and 5,430 subjects undergoing Traditional colonoscopy. The AI-assisted endoscopy performed better in comparison to traditional colonoscopy in detecting adenomas (OR = 1.45 (1.27;1.67) ,95% CI, P = 0.003, I2 = 63.66%). The detection rates of detecting polyps in AI-assisted colonoscopy were significantly higher than traditional colonoscopy (OR = 2.01(1.14;3.52) 95% CI, P < 0.0001, I2 = 94.6%). Conclusion: Our study shows that AI assisted endoscopy is superior to traditional colonoscopy in the detection of adenomas. The detection rate of polyps through AI assisted colonoscopy was significantly higher. This highlights the importance of integration of AI in routine endoscopy to enhance detection of adenomas and polyps which go undetected through traditional endoscopy.

Authors

Publication Details

Journal
The American Journal of Gastroenterology
Published
2025-10-01
DOI
https://doi.org/10.14309/01.ajg.0001129772.89523.94
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

S578 Comparison of Effectiveness of Artificial Intelligence (AI)-Aided Colonoscopy With Conventional Colonoscopy in Enhanced Detection of Colorectal Adenomas and Polyps—A Systematic Review and Meta-Analysis

Omar Oudit, Ananda S. Prasad, Sravani Bhavanam, Adithya Sathya Narayana et al.
The American Journal of Gastroenterology
Radiomics and Machine Learning in Medical Imaging
article

S578 Comparison of Effectiveness of Artificial Intelligence (AI)-Aided Colonoscopy With Conventional Colonoscopy in Enhanced Detection of Colorectal Adenomas and Polyps—A Systematic Review and Meta-Analysis

Omar Oudit, Ananda S. Prasad, Sravani Bhavanam, Adithya Sathya Narayana, Veena R. Iyer, Tamilselvan Mohan, Prakriti Ramamurthy, Druvadeep Srinivas, Vinay Chandramouli Bellur, Rishikesh R. Magaji, Vibhav MS
article en

Abstract

Introduction: Artificial intelligence (AI)-aided colonscopy has shown significant improvement in detection of precancerous lesions. This systematic review and meta-analysis evaluates its performance in comparison to regular conventional colonoscopy by assessing the adenoma and polyp detection rates, as well as adenomas and polyps detected per procedure, thus highlighting the potential of AI to enhance prevention of colorectal cancer. Methods: The review conducted follows the PRISMA guidelines and major medical databases, which include PUBMED, Google Scholar and Science-Direct, were extensively searched using a comprehensive search term to identify and retrieve available articles. The articles that included the assessment of the detection of adenoma and polyp by AI-assisted colonoscopy and traditional colonoscopy were included in the final analysis. The data was analyzed using the Meta, Metadata and the Metafor packages of R Studio. The odds ratio (OR) of detection of adenoma and polyp through different colonoscopy techniques was assessed as the primary outcome. The Mantel-Haenszel method and the Inverse variance method were utilized to analyze the odds ratio. The I2 test was used to assess the heterogeneity of the studies. Results: The study included a total of 16 studies with 5,466 patients undergoing AI-assisted colonoscopy and 5,430 subjects undergoing Traditional colonoscopy. The AI-assisted endoscopy performed better in comparison to traditional colonoscopy in detecting adenomas (OR = 1.45 (1.27;1.67) ,95% CI, P = 0.003, I2 = 63.66%). The detection rates of detecting polyps in AI-assisted colonoscopy were significantly higher than traditional colonoscopy (OR = 2.01(1.14;3.52) 95% CI, P < 0.0001, I2 = 94.6%). Conclusion: Our study shows that AI assisted endoscopy is superior to traditional colonoscopy in the detection of adenomas. The detection rate of polyps through AI assisted colonoscopy was significantly higher. This highlights the importance of integration of AI in routine endoscopy to enhance detection of adenomas and polyps which go undetected through traditional endoscopy.

The American Journal of GastroenterologyVol. 120(10S2)
Openalex Percentile: Top 55%
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.