Abstract
Gastroesophageal reflux disease (GERD) is a common disease with high prevalence, and its endoscopic severity can be evaluated using the Los Angeles classification (LA grade). This paper proposes a deep learning model (i.e., GERD-VGGNet) that employs convolutional neural networks for automatic classification and interpretation of routine GERD LA grade. The proposed model employs a data augmentation technique, a two-stage no-freezing fine-tuning policy, and an early stopping criterion. As a result, the proposed model exhibits high generalizability. A dataset of images from 464 patients was used for model training and validation. An additional 32 patients served as a test set to evaluate the accuracy of both the model and our trainees. Experimental results demonstrate that the best model for the development set exhibited an overall accuracy of 99.2% (grade A–B), 100% (grade C–D), and 100% (normal group) using narrow-band image (NBI) endoscopy. On the test set, the proposed model resulted in an accuracy of 87.9%, which was significantly higher than the results of the trainees (75.0% and 65.6%). The proposed GERD-VGGNet model can assist automatic classification of GERD in conventional and NBI environments and thereby increase the accuracy of interpretation of the results by inexperienced endoscopists.
Highlights
Gastroesophageal reflux disease (GERD), which is a condition that develops when the reflux of stomach contents causes symptoms of discomfort and/or associated complications [1], is among the diseases with the highest prevalence over the past two decades [2,3].GERD can be classified as either erosive or non-erosive esophagitis and is characterized by endoscopically visible breaks in the distal esophageal mucosa in the former category and a lack of such breaks in the latter [4,5]
The results demonstrate that seven imagers were misclassified by the proposed GERD-VGGNet), 10 images were misclassified by trainee 1, and seven images were misclassified by trainee 2
The experimental results clearly demonstrate that data augmentation is key to training a deep learning neural network
Summary
Gastroesophageal reflux disease (GERD), which is a condition that develops when the reflux of stomach contents causes symptoms of discomfort and/or associated complications [1], is among the diseases with the highest prevalence over the past two decades [2,3]. GERD can be classified as either erosive or non-erosive esophagitis and is characterized by endoscopically visible breaks in the distal esophageal mucosa in the former category and a lack of such breaks in the latter [4,5]. Double-contrast barium swallow examination has been previously used to diagnose GERD [6], esophagogastroduodenoscopy (EGD). The popular and powerful Los Angeles classification (LA grade) system, which was established more than 20 years ago, is used in endoscopy examinations to classify GERD [9].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callMore From: International Journal of Environmental Research and Public Health
Disclaimer: 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.
