Early Barrett’s Neoplasia Detection

This study investigated the feasibility of a computer algorithm to detect early neoplasia in Barrett’s esophagus (BE) using ex vivo volumetric laser endomicroscopy (VLE) images. The algorithm, which utilized features such as layering and signal decay statistics, showed strong performance, outperforming the traditional clinical VLE prediction score. The results indicate that an automated detection algorithm could effectively assist endoscopists in real-time detection of early neoplasia, with further validation required in in vivo VLE scans.