Optical Diagnosis of Colorectal Polyps: Recent Developments.

Autor: Djinbachian, Roupen, Dubé, Anne-Julie, von Renteln, Daniel
Zdroj: Current Treatment Options in Gastroenterology; Mar2019, Vol. 17 Issue 1, p99-114, 16p
Abstrakt: Purpose of review: Optical diagnosis of diminutive colorectal polyps has been recently proposed as an alternative to histopathologic diagnosis. Recent developments in imaging techniques, new classification systems, and the use of artificial intelligence have allowed for increased viability of optical diagnosis. This review provides an up-to-date overview of optical diagnosis recommendations, classifications, outcomes, and recent developments.Recent findings: There are currently seven major classification systems and three major society recommendations for quality benchmarks for optical diagnosis of diminutive polyps. The NICE classification has been extensively studied and meets quality benchmarks for most imaging techniques but does not allow for the diagnosis of sessile serrated polyps (SSPs). The SIMPLE classification has met quality benchmarks for NBI and i-Scan and allows for the diagnosis of SSPs. Other classification systems need to be further studied to validate effectiveness. Computer-assisted diagnosis of colorectal polyps is a very promising recent development with first studies showing that society-recommended quality benchmarks for real-time colonoscopies on patients are being met. Limitations include a non-negligible percentage of failure to diagnose, low specificity, and low number of real-time diagnostic studies. More research needs to be performed to further understand the value of artificial intelligence for optical polyp diagnosis.Summary: Optical diagnosis of diminutive colorectal polyps is currently a viable strategy for experienced endoscopists using validated classifications and imaging-enhanced endoscopy. Artificial intelligence-based diagnosis could make optical diagnosis widely applicable but is currently in its early developmental stage. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index