Endoscopy-Driven Pretraining for Classification of Dysplasia in Barrett’s Esophagus with Endoscopic Narrow-Band Imaging Zoom Videos

• Main Authors: Joost van der Putten, Maarten Struyvenberg, Jeroen de Groof, Wouter Curvers, Erik Schoon, Francisco Baldaque-Silva, Jacques Bergman, Fons van der Sommen, Peter H.N. de With
• Format: Article
• Language: English
• Published: MDPI AG 2020-05-01
• Series: Applied Sciences
• Subjects: endoscopic zoom imagery; Barrett’s esophagus; deep learning; classification; machine learning
• Online Access: https://www.mdpi.com/2076-3417/10/10/3407

Endoscopic diagnosis of early neoplasia in Barrett’s Esophagus is generally a two-step process of primary detection in overview, followed by detailed inspection of any visible abnormalities using Narrow Band Imaging (NBI). However, endoscopists struggle with evaluating NBI-zoom imagery of subtle abnormalities. In this work, we propose the first results of a deep learning system for the characterization of NBI-zoom imagery of Barrett’s Esophagus with an accuracy, sensitivity, and specificity of 83.6%, 83.1%, and 84.0%, respectively. We also show that endoscopy-driven pretraining outperforms two models, one without pretraining as well as a model with ImageNet initialization. The final model outperforms absence of pretraining by approximately 10% and the performance is 2% higher in terms of accuracy compared to ImageNet pretraining. Furthermore, the practical deployment of our model is not hampered by ImageNet licensing, thereby paving the way for clinical application.