JMIR Publications

ABSTRACT

Background:

The majority of colorectal polyps are diminutive and benign, especially for those in the rectosigmoid colon, and resecting these polyps is not cost-effective. Advancements in image-enhanced endoscopy has improved the optical prediction of histology in colorectal polyps. However, subjective interpretability and inter-/intra-observer variability prohibited the widespread implementation. Studies on computer-aided diagnosis (CAD) are increasing; however, their small sample size limits the statistical significance.

Objective:

to evaluate the diagnostic test accuracy of CAD models in predicting the histology of diminutive colorectal polyps using endoscopic images.

Methods:

Core databases were searched for studies based on endoscopic imaging using CAD models for the histologic diagnosis of diminutive colorectal polyps and presenting data on diagnostic performance. Systematic review and diagnostic test accuracy meta-analysis were performed.

Results:

Overall, 13 studies were included. The pooled area under the curve, sensitivity, specificity, and diagnostic odds ratio of CAD models for the diagnosis of diminutive colorectal polyps (adenomatous or neoplastic vs. nonadenomatous or nonneoplastic) were 0.96 (95% confidence interval, 0.93–0.97), 0.93 (0.91–0.95), 0.87 (0.76–0.93), and 87 (38–201), respectively. Meta-regression showed no heterogeneity, and no publication bias was detected. Subgroup analyses showed robust results. The negative predictive value of CAD models for the diagnosis of adenomatous polyp in the rectosigmoid colon was 0.96 (0.95–0.97), exceeding the threshold of the “diagnosis and leave” strategy.

Conclusions:

CAD models show potential for the optical histologic diagnosis of diminutive colorectal polyps using endoscopic images.