Cargando…
An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks
Generative adversarial networks (GAN) in medicine are valuable techniques for augmenting unbalanced rare data, anomaly detection, and avoiding patient privacy issues. However, there were limits to generating high-quality endoscopic images with various characteristics, such as peristalsis, viewpoints...
Autores principales: | , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Springer International Publishing
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10406771/ https://www.ncbi.nlm.nih.gov/pubmed/36914855 http://dx.doi.org/10.1007/s10278-023-00803-2 |
_version_ | 1785085812927365120 |
---|---|
author | Shin, Keewon Lee, Jung Su Lee, Ji Young Lee, Hyunsu Kim, Jeongseok Byeon, Jeong-Sik Jung, Hwoon-Yong Kim, Do Hoon Kim, Namkug |
author_facet | Shin, Keewon Lee, Jung Su Lee, Ji Young Lee, Hyunsu Kim, Jeongseok Byeon, Jeong-Sik Jung, Hwoon-Yong Kim, Do Hoon Kim, Namkug |
author_sort | Shin, Keewon |
collection | PubMed |
description | Generative adversarial networks (GAN) in medicine are valuable techniques for augmenting unbalanced rare data, anomaly detection, and avoiding patient privacy issues. However, there were limits to generating high-quality endoscopic images with various characteristics, such as peristalsis, viewpoints, light sources, and mucous patterns. This study used the progressive growing of GAN (PGGAN) within the normal distribution dataset to confirm the ability to generate high-quality gastrointestinal images and investigated what barriers PGGAN has to generate endoscopic images. We trained the PGGAN with 107,060 gastroscopy images from 4165 normal patients to generate highly realistic 512(2) pixel-sized images. For the evaluation, visual Turing tests were conducted on 100 real and 100 synthetic images to distinguish the authenticity of images by 19 endoscopists. The endoscopists were divided into three groups based on their years of clinical experience for subgroup analysis. The overall accuracy, sensitivity, and specificity of the 19 endoscopist groups were 61.3%, 70.3%, and 52.4%, respectively. The mean accuracy of the three endoscopist groups was 62.4 [Group I], 59.8 [Group II], and 59.1% [Group III], which was not considered a significant difference. There were no statistically significant differences in the location of the stomach. However, the real images with the anatomical landmark pylorus had higher detection sensitivity. The images generated by PGGAN showed highly realistic depictions that were difficult to distinguish, regardless of their expertise as endoscopists. However, it was necessary to establish GANs that could better represent the rugal folds and mucous membrane texture. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10278-023-00803-2. |
format | Online Article Text |
id | pubmed-10406771 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Springer International Publishing |
record_format | MEDLINE/PubMed |
spelling | pubmed-104067712023-08-09 An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks Shin, Keewon Lee, Jung Su Lee, Ji Young Lee, Hyunsu Kim, Jeongseok Byeon, Jeong-Sik Jung, Hwoon-Yong Kim, Do Hoon Kim, Namkug J Digit Imaging Article Generative adversarial networks (GAN) in medicine are valuable techniques for augmenting unbalanced rare data, anomaly detection, and avoiding patient privacy issues. However, there were limits to generating high-quality endoscopic images with various characteristics, such as peristalsis, viewpoints, light sources, and mucous patterns. This study used the progressive growing of GAN (PGGAN) within the normal distribution dataset to confirm the ability to generate high-quality gastrointestinal images and investigated what barriers PGGAN has to generate endoscopic images. We trained the PGGAN with 107,060 gastroscopy images from 4165 normal patients to generate highly realistic 512(2) pixel-sized images. For the evaluation, visual Turing tests were conducted on 100 real and 100 synthetic images to distinguish the authenticity of images by 19 endoscopists. The endoscopists were divided into three groups based on their years of clinical experience for subgroup analysis. The overall accuracy, sensitivity, and specificity of the 19 endoscopist groups were 61.3%, 70.3%, and 52.4%, respectively. The mean accuracy of the three endoscopist groups was 62.4 [Group I], 59.8 [Group II], and 59.1% [Group III], which was not considered a significant difference. There were no statistically significant differences in the location of the stomach. However, the real images with the anatomical landmark pylorus had higher detection sensitivity. The images generated by PGGAN showed highly realistic depictions that were difficult to distinguish, regardless of their expertise as endoscopists. However, it was necessary to establish GANs that could better represent the rugal folds and mucous membrane texture. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10278-023-00803-2. Springer International Publishing 2023-03-13 2023-08 /pmc/articles/PMC10406771/ /pubmed/36914855 http://dx.doi.org/10.1007/s10278-023-00803-2 Text en © The Author(s) 2023 https://creativecommons.org/licenses/by/4.0/Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) . |
spellingShingle | Article Shin, Keewon Lee, Jung Su Lee, Ji Young Lee, Hyunsu Kim, Jeongseok Byeon, Jeong-Sik Jung, Hwoon-Yong Kim, Do Hoon Kim, Namkug An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title | An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title_full | An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title_fullStr | An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title_full_unstemmed | An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title_short | An Image Turing Test on Realistic Gastroscopy Images Generated by Using the Progressive Growing of Generative Adversarial Networks |
title_sort | image turing test on realistic gastroscopy images generated by using the progressive growing of generative adversarial networks |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10406771/ https://www.ncbi.nlm.nih.gov/pubmed/36914855 http://dx.doi.org/10.1007/s10278-023-00803-2 |
work_keys_str_mv | AT shinkeewon animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leejungsu animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leejiyoung animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leehyunsu animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimjeongseok animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT byeonjeongsik animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT junghwoonyong animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimdohoon animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimnamkug animageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT shinkeewon imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leejungsu imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leejiyoung imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT leehyunsu imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimjeongseok imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT byeonjeongsik imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT junghwoonyong imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimdohoon imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks AT kimnamkug imageturingtestonrealisticgastroscopyimagesgeneratedbyusingtheprogressivegrowingofgenerativeadversarialnetworks |