Time-saving polyp detection in colon capsule endoscopy: evaluation of a novel software algorithm

Abstract

Background: Colon capsule endoscopy (CCE) is a reliable method to detect colonic polyps in the well-prepared colon. As CCE evaluation can be time consuming, a new software algorithm might aid in reducing evaluation time.

Objectives: The aim of the study was to evaluate whether it is feasible to reliably detect colon polyps in CCE videos with a new software algorithm the "collage mode" (Rapid 8 Software, Covidien/Medtronic®).

Methods: Twenty-nine CCE videos were randomly presented to three experienced and to three inexperienced investigators. Videos were evaluated by applying the collage mode. Investigation time was documented and the results (≥one polyp vs. no polyp) were compared with the findings of two highly experienced central readers who read the CCE videos in the standard mode beforehand.

Results: It took a median time of 9.8, 3.5, and 7.5 vs. 4.3, 4.6 and 12.5 min for experienced vs. inexperienced investigators to review the CCE videos. For detecting ≥one polyp vs. no polyp, sensitivity of 93.3%, 73.3%, and 93.3% was observed for the experienced and sensitivity of 46.7%, 33.3%, and 93.3% for the inexperienced CCE readers.

Conclusion: Collage mode might allow for a quick review of CCE videos with a high polyp detection rate for experienced CCE readers. Future prospective studies should include CCE collage mode for rapid polyp detection to further prove the feasibility of practical colon polyp detection by CCE and possibly support the role of CCE as a screening tool in CRC prevention.

Keywords: Adenoma; Capsule endoscopy; Colon capsule endoscopy; Colonoscopy; Endoscopy; Polyp.