Related Experiment Video
Updated: Oct 15, 2025
07:36
Laparoscopic Common Bile Duct Exploration in Patients with a Previous History of Biliary Tract Surgery
Published on: February 10, 2023
2.0K
Automatic Identification of Papillary Projections in Indeterminate Biliary Strictures Using Digital Single-Operator
Tiago Ribeiro1,2, Miguel Mascarenhas Saraiva1,2,3, João Afonso1,2
1Department of Gastroenterology, São João University Hospital, Alameda Professor Hernâni Monteiro, Porto, Portugal.
Clinical and Translational Gastroenterology
|October 27, 2021
Summary
An artificial intelligence (AI) model using a convolutional neural network (CNN) accurately detects papillary projections (PP) in cholangioscopy images. This AI tool shows promise for improving the characterization of biliary strictures, aiding in malignancy assessment.
Area of Science:
- Gastroenterology
- Medical Imaging
- Artificial Intelligence
Background:
- Characterizing biliary strictures is complex, with papillary projections (PP) indicating high malignancy risk during cholangioscopy.
- Artificial intelligence (AI) is increasingly studied for endoscopic applications.
Purpose of the Study:
- To develop an AI algorithm for the automatic detection of papillary projections (PP) in digital single-operator cholangioscopy images.
Main Methods:
- A convolutional neural network (CNN) was developed and trained on 3,920 images from 85 patients.
- The CNN's performance was evaluated using area under the curve (AUC), sensitivity, specificity, and predictive values.
Main Results:
- The AI model achieved high accuracy, with a sensitivity of 99.7% and specificity of 97.1%.
- The area under the curve (AUC) for the CNN model was 1.00, indicating excellent performance.
Conclusions:
- The developed CNN effectively detects papillary projections (PP) in cholangioscopy images with high accuracy.
- Future AI tools could enhance the macroscopic characterization of biliary strictures.
