Algorithm combining virtual chromoendoscopy features for colorectal polyp classification.

Background and study aims Colonoscopy is considered the gold standard for decreasing colorectal cancer incidence and mortality. Optical diagnosis of colorectal polyps (CRPs) is an ongoing challenge in clinical colonoscopy and its accuracy among endoscopists varies widely. Computer-aided diagnosis (CAD) for CRP characterization may help to improve this accuracy. In this study, we investigated the diagnostic accuracy of a novel algorithm for polyp malignancy classification by exploiting the complementary information revealed by three specific modalities. Methods We developed a CAD algorithm for CRP characterization based on high-definition, non-magnified white light (HDWL), Blue light imaging (BLI) and linked color imaging (LCI) still images from routine exams. All CRPs were collected prospectively and classified into benign or premalignant using histopathology as gold standard. Images and data were used to train the CAD algorithm using triplet network architecture. Our training dataset was validated using a threefold cross validation. Results In total 609 colonoscopy images of 203 CRPs of 154 consecutive patients were collected. A total of 174 CRPs were found to be premalignant and 29 were benign. Combining the triplet network features with all three image enhancement modalities resulted in an accuracy of 90.6 %, 89.7 % sensitivity, 96.6 % specificity, a positive predictive value of 99.4 %, and a negative predictive value of 60.9 % for CRP malignancy classification. The classification time for our CAD algorithm was approximately 90 ms per image. Conclusions Our novel approach and algorithm for CRP classification differentiates accurately between benign and premalignant polyps in non-magnified endoscopic images. This is the first algorithm combining three optical modalities (HDWL/BLI/LCI) exploiting the triplet network approach.

Interval cancers after colonoscopy-insights and recommendations.

Several studies have raised warnings about the limited effectiveness of colonoscopy for the prevention of colorectal cancer (CRC), especially of the proximal colon. Two major categories of factors might be responsible for the development of interval cancers, namely technical, endoscopist-dependent factors and biological characteristics of the cancer that lead to more rapid tumour progression. Recognition of endoscopist-dependent factors is critical, as these factors are probably amenable to correction through improved awareness and education of endoscopists, using quality metrics (such as adenoma detection rates and cecal intubation rates) for objective evaluation and feedback. In this article, the current literature regarding the incidence of, and potential explanations for, interval CRCs is outlined. Although there is probably an interaction between technical and biology-related factors--and an attempt to dissect the biology from the technology might be fraught with difficulties--a structured analysis of individual cases of interval cancer might help in the continuous monitoring of the quality of colonoscopy, and ultimately might reduce the number of interval CRCs.