Endoscopy 2023; 55(08): v23
DOI: 10.1055/a-2073-5254
Author commentary

Author commentary on Carlos Robles-Medranda et al.

Carlos Robles-Medranda et al. Artificial intelligence for diagnosing neoplasia on digital cholangioscopy: development and multicenter validation of a convolutional neural network model

The authors developed an artificial intelligence model to recognize macroscopic morphological characteristics of neoplastic lesions during real-time digital cholangioscopy. When applied to cholangioscopy video recordings from 170 patients, the model achieved sensitivity and specificity for diagnosis of neoplasia of 91 % and 68 %, respectively, and either matched or outperformed expert endoscopists.


Publication History

Article published online:
27 July 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany