Automatic detection of tumor vessels in indeterminate biliary strictures in digital single-operator cholangioscopy.

Digital cholangioscopy has become an essential tool for the assessment of biliary strictures and is essential to predict its malignant potential.

Our group developed a deep learning algorithm based on convolutional neural networks for automatic detection of tumor vessels in digital cholangioscopy images (Spyglass DS II). Our system detected this cholangioscopic feature with a sensitivity and specificity of 99%.The potential improvement in the accuracy of visual diagnosis of indeterminate biliary strictures may provide significant outcome improvements.

The link for the complete article can be found here.

In Press // March 2022