In Vivo Computer-aided Prediction of Polyp Histology on White Light Colonoscopy

In a previous study, our group developed a computer assisted system (CAD) based on the extraction of polyp surface patterns using high-definition white light (WL-HD). This CAD was able to distinguish between dysplastic and non-dysplastic polyps with an accuracy> 90%. The current goal is to validate an improved version of this system in routine clinical practice. Since both the American Society of Gastrointestinal Endoscopy (ASGE) and the European Society of Gastrointestinal Endoscopy (ESGE) recommend performing optical diagnosis only by expert professionals and for diminutive polyps, the main objective of our study will be to assess the accuracy of the method for polyps of small size in real life during colonoscopies performed by endoscopists with different levels of expertise in optical diagnosis. The performance of the CAD method will be compared with the in-situ histological prediction made by the endoscopists and it will be recorded if their decision changes based on the system prediction. Histological analysis will be the gold standard.

In Vivo Computer-aided Prediction of Polyp Histology on White Light Colonoscopy

Recruitment

Summary

Participation Requirements

Description

Tracking Information