Characterization of Mucosal Lesions in Crohn's Disease Scored With Capsule Endoscopy: A Systematic Review

Abstract

Background and Aims: There is little agreement on the nomenclature and description of Crohn's disease (CD) lesions that can be found in the small and large bowel using capsule endoscopy (CE). We performed a systematic review to identify mucosal lesions that have been described using CE in CD, in both the small bowel and colon, with the aim to make propositions to homogenize such descriptions.Methods: A systematic literature search was conducted using Embase, Medline (OvidSP), and Cochrane Central on August 6, 2019. Clinical studies providing nomenclature and descriptions for small bowel and colonic inflammatory lesions using CE in CD were selected for data collection.Results: In total, 851 articles were included for abstract screening out of which 219 were analyzed for full-text review. Twenty-two articles were selected for data extraction. Seven items, accompanied by clear descriptions, were found for the small bowel: i.e., ulcer, erosion, aphthoid lesion, edema, fissure, cobblestone appearance, and villous atrophy. No studies were found describing inflammatory items using CE in colonic CD.Conclusions: The most frequently described CD lesions using CE were ulcers and erosions. Subjective interpretation of CE inflammatory findings plays an important role. Based on our findings, a range of suggestions regarding items and descriptions is made that might form the basis of a pan-enteric CE activity index.

Automated Bowel Polyp Detection Based on Actively Controlled Capsule Endoscopy: Feasibility Study

“…These systems are built up from the supervised machine learning and deep learning algorithms that have been trained on the dataset image samples taken from the CE camera with some gold-standard features of normal and abnormal objects such as color, texture, or shape [33,34]. These systems have demonstrated promising results with high sensitivity and specificity for obscure gastrointestinal bleeding (OGIB) [35][36][37], Crohn disease lesions [38,39], polyps [40][41][42], and so on. Especially for colorectal polyp detection, deep learning models show remarkable performances as summarized in [43].…”