Synthesizing realistic high-resolution retina image by style-based generative adversarial network and its utilization
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Mingyu | - |
dc.contributor.author | Kim, You Na | - |
dc.contributor.author | Jang, Miso | - |
dc.contributor.author | Hwang, Jeongeun | - |
dc.contributor.author | Kim, Hong-Kyu | - |
dc.contributor.author | Yoon, Sang Chul | - |
dc.contributor.author | Kim, Yoon Jeon | - |
dc.contributor.author | Kim, Namkug | - |
dc.date.accessioned | 2022-11-08T01:40:30Z | - |
dc.date.available | 2022-11-08T01:40:30Z | - |
dc.date.issued | 2022-10 | - |
dc.identifier.issn | 2045-2322 | - |
dc.identifier.issn | 2045-2322 | - |
dc.identifier.uri | https://scholarworks.korea.ac.kr/kumedicine/handle/2021.sw.kumedicine/61724 | - |
dc.description.abstract | Realistic image synthesis based on deep learning is an invaluable technique for developing high-performance computer aided diagnosis systems while protecting patient privacy. However, training a generative adversarial network (GAN) for image synthesis remains challenging because of the large amounts of data required for training various kinds of image features. This study aims to synthesize retinal images indistinguishable from real images and evaluate the efficacy of the synthesized images having a specific disease for augmenting class imbalanced datasets. The synthesized images were validated via image Turing tests, qualitative analysis by retinal specialists, and quantitative analyses on amounts and signal-to-noise ratios of vessels. The efficacy of synthesized images was verified by deep learning-based classification performance. Turing test shows that accuracy, sensitivity, and specificity of 54.0 +/- 12.3%, 71.1 +/- 18.8%, and 36.9 +/- 25.5%, respectively. Here, sensitivity represents correctness to find real images among real datasets. Vessel amounts and average SNR comparisons show 0.43% and 1.5% difference between real and synthesized images. The classification performance after augmenting synthesized images outperforms every ratio of imbalanced real datasets. Our study shows the realistic retina images were successfully generated with insignificant differences between the real and synthesized images and shows great potential for practical applications. | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | Nature Publishing Group | - |
dc.title | Synthesizing realistic high-resolution retina image by style-based generative adversarial network and its utilization | - |
dc.type | Article | - |
dc.publisher.location | 영국 | - |
dc.identifier.doi | 10.1038/s41598-022-20698-3 | - |
dc.identifier.scopusid | 2-s2.0-85139888168 | - |
dc.identifier.wosid | 000869292100018 | - |
dc.identifier.bibliographicCitation | Scientific Reports, v.12, no.1 | - |
dc.citation.title | Scientific Reports | - |
dc.citation.volume | 12 | - |
dc.citation.number | 1 | - |
dc.type.docType | Article | - |
dc.description.isOpenAccess | Y | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Science & Technology - Other Topics | - |
dc.relation.journalWebOfScienceCategory | Multidisciplinary Sciences | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
73, Goryeodae-ro, Seongbuk-gu, Seoul, Republic of Korea (02841)82-2-2286-1265
COPYRIGHT 2020 KOREA UNIVERSITY MEDICAL LIBRARY ALL RIGHTS RESERVED.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.