Barrett’s esophagus (BE) is a known precursor for esophageal adenocarcinoma (EAC). Therefore, BE patients undergo regular surveillance with white-light endoscopy and quadrantic random biopsies. However, the current surveillance protocol is not efficient since early cancer is often missed due to subtle appearance upon endoscopy and sampling error. Volumetric Laser Endomicroscopy (VLE) is a new balloon-based system, that provides a 3D scan of the Barrett’s segment during endoscopy. This enables evaluation of the entire Barrett’s segment including subsurface layers, leading to better detection of early cancer. In addition, with the new VLE laser marking tool, suspicious areas can be directly marked and sampled. Due to the amount of data and subtle gray shades, interpretation of VLE scans by the human eye seems complex. Therefore, automated image analysis algorithms may be of great additional value to aid in detection of BE neoplasia on VLE scans.
Objectives: The aim of this study is to evaluate and validate a clinical VLE prediction score and a VLE computer algorithm for the detection of BE neoplasia.
