Abstract
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.
Highlights
Barrett’s Esophagus (BE) is a known precursor for Esophageal AdenoCarcinoma (EAC), a form of gastrointestinal cancer with a poor prognosis
The aim of the current study is to investigate the feasibility of a deep learning algorithm for the characterization of Narrow Band Imaging (NBI)-zoom imagery after primary detection in White Light Endoscopy (WLE)
We initially explored extracting the additional information contained in multi-modal imaging, using deep learning and a combination of NBI and WLE in overview, to improve localization scores for dysplasia in BE [13]
Summary
Barrett’s Esophagus (BE) is a known precursor for Esophageal AdenoCarcinoma (EAC), a form of gastrointestinal cancer with a poor prognosis. For this reason, patients with BE undergo regular endoscopic surveillance. Patients with BE undergo regular endoscopic surveillance This offers early detection of neoplasia and enables endoscopic treatment, which has an excellent prognosis [1,2]. The current BE surveillance protocol is suboptimal, mainly due to the fact that neoplasia is difficult to detect because of their subtle endoscopic appearances [1]. BE surveillance is time consuming, expensive, and has a high potential for sampling error of random biopsies [3]. Primary detection of a suspected lesion is determined with White Light Endoscopy (WLE) in overview, followed by a detailed inspection of the mucosa for any visible abnormalities and to identify the lesion boundaries
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
