Department of Convergence Medicine, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea.
Department of Ophthalmology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, 05505, Republic of Korea.
Sci Rep. 2022 Oct 15;12(1):17307. doi: 10.1038/s41598-022-20698-3.
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.
基于深度学习的真实感图像合成是开发高性能计算机辅助诊断系统的宝贵技术，同时还能保护患者隐私。然而，由于训练各种图像特征需要大量数据，因此训练生成对抗网络 (GAN) 进行图像合成仍然具有挑战性。本研究旨在合成与真实图像难以区分的视网膜图像，并评估具有特定疾病的合成图像增强类不平衡数据集的效果。通过图像图灵测试、视网膜专家的定性分析以及对血管数量和信噪比的定量分析来验证合成图像的效果。通过基于深度学习的分类性能来验证合成图像的效果。图灵测试的结果表明，准确率、敏感度和特异性分别为 54.0±12.3%、71.1±18.8%和 36.9±25.5%。这里，敏感度表示在真实数据集找到真实图像的正确率。血管数量和平均信噪比的比较表明，真实图像和合成图像之间的差异为 0.43%和 1.5%。在增强合成图像后，分类性能优于每个不平衡的真实数据集的比例。我们的研究表明，已经成功生成了具有真实感的视网膜图像，且真实图像和合成图像之间几乎没有差异，这为实际应用提供了巨大的潜力。
