Prospective development and validation of a volumetric laser endomicroscopy computer algorithm for detection of Barrett’s neoplasia
Volumetric laser endomicroscopy is used to detect Barrett's esophagus (BE) dysplasia. Real-time interpretation of VLE scans is complex and time-consuming. VLE PREDICT study prospectively enrolled 47 patients with Barrett's esophagus (BE) dysplasia. In total, 229 non-dysplastic BE and 89 neoplastic targets were laser marked. Deep convolutional neural networks were used to construct a CAD algorithm. The algorithm detected neoplasia with high accuracy and outperformed 10 VLE experts.
