Previously submitted to: Journal of Medical Internet Research (no longer under consideration since Jan 25, 2022)
Date Submitted: Nov 25, 2020
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
An Image Turing Test on Realistic Gastroscopy Images Generated by using the Progressive Growing of Generative Adversarial Networks
ABSTRACT
Background:
An Image Turing Test on Realistic Gastroscopy Images Generated by using the Progressive Growing of Generative Adversarial Networks(PGGAN)
Objective:
This study aims to present a method for generating gastroscopy images using a deep learning-based progressive growing of generative adversarial networks (PGGAN) as the first step for anomaly detection.
Methods:
We trained the PGGAN with a total 107,060 normal gastroscopy images to generate highly realistic images 512 x 512 pixels in size. For the evaluation, image Turing tests were conducted on 200 images, including 100 real and 100 synthesized images, by 19 endoscopists. The endoscopists were divided into three groups based on their years of clinical experience (0 to 5, 5 to 10, and 10 or more).
Results:
For the image Turing test, the mean accuracy, sensitivity, and specificity of the 19 endoscopists were 61.3%, 70.3%, and 52.4%, respectively. The mean accuracy of the three endoscopist groups was 62.4 (0 to 5 yrs.), 59.8 (5 to 10 yrs.), and 59.1 % (10 or more yrs.), which was not considered a significant difference. There were no statistically significant differences in the location of the stomach. However, in sensitivity to anatomical landmarks, the sensitivity to the pylorus was higher (P = 0.002).
Conclusions:
Images generated by PGGAN showed highly realistic depictions that were difficult to distinguish, regardless of expertise, and could be used for anomaly detection in the future.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.