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 5122 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.
O(on behalf of the Intestinal Tumor Research Group of the Korean Association for the Study of Intestinal Diseases (KASID)
on behalf of the Intestinal Tumor Research Group of the Korean Association for the Study of Intestinal Diseases (KASID)
+12
Background/Aims: We investigated the clinical practice patterns of post-polypectomy colonoscopic surveillance among Korean endoscopists.Methods: In a web-based survey conducted between September and November 2021, participants were asked about their preferred surveillance intervals and the patient age at which surveillance was discontinued. Adherence to the recent guidelines of the U.S. Multi-Society Task Force on Colorectal Cancer (USMSTF) was also analyzed.Results: In total, 196 endoscopists completed the survey. The most preferred first surveillance intervals were: a 5-year interval after the removal of 1–2 tubular adenomas < 10 mm; a 3-year interval after the removal of 3–10 tubular adenomas < 10 mm, adenomas ≥ 10 mm, tubulovillous or villous adenomas, ≤ 20 hyperplastic polyps < 10 mm, 1–4 sessile serrated lesions (SSLs) < 10 mm, hyperplastic polyps or SSLs ≥ 10 mm, and traditional serrated adenomas; and a 1-year interval after the removal of adenomas with highgrade dysplasia, >10 adenomas, 5–10 SSLs, and SSLs with dysplasia. In piecemeal resections of large polyps ( > 20 mm), surveillance colonoscopy was mostly preferred after 1 year for adenomas and 6 months for SSLs. The mean USMSTF guideline adherence rate was 30.7%. The largest proportion of respondents (40.8%–55.1%) discontinued the surveillance at the patient age of 80–84 years.Conclusions: A significant discrepancy was observed between the preferred post-polypectomy surveillance intervals and recent international guidelines. Individualized measures are required to increase adherence to the guidelines.
Ji Eun Baek, Hahn Yi, Seung Wook Hong, Subin Song, Ji Young Lee, Sung Wook Hwang, Sang Hyoung Park, Dong-Hoon Yang, Byong Duk Ye, Seung-Jae Myung, Suk-Kyun Yang, Namkug Kim, Jeong-Sik Byeon. Gut and Liver 2025;19:69-76. https://doi.org/10.5009/gnl240273
I(Intestinal Tumor Research Group of the Korean Association for the Study of Intestinal Diseases (KASID)
Intestinal Tumor Research Group of the Korean Association for the Study of Intestinal Diseases (KASID)
+12
Jeongseok Kim, Tae-Geun Gweon, Min Seob Kwak, Su Young Kim, Seong Jung Kim, Hyun Gun Kim, Sung Noh Hong, Eun Sun Kim, Chang Mo Moon, Dae Seong Myung, Dong-Hoon Baek, Shin Ju Oh, Hyun Jung Lee, Ji Young Lee, Yunho Jung, Jaeyoung Chun, Dong-Hoon Yang, Eun Ran Kim, Intestinal Tumor Research Group of the Korean Association for the Study of Intestinal Diseases (KASID). Gut and Liver -0001;0:. https://doi.org/10.5009/gnl240217
Hana Park, Ji Young Lee, Sungwon Park, Hyo Jeong Lee, Suh Eun Bae, Jaeil Kim, Hye-Sook Chang, Jaewon Choe, Hye Won Park, Ju Hyun Shim. Gut and Liver -0001;0:. https://doi.org/10.5009/gnl250053