From the Department of Applied Mathematics, Faculty of Electrical Engineering, Mathematics and Computer Science, Technical Medical Centre, University of Twente, Zilverling, PO Box 217, 7500 AE Enschede, the Netherlands (J.M.W.); Department of Biomedical Engineering and Physics (J.M.W., I.I.) and Department of Radiology and Nuclear Medicine (I.I.), Amsterdam University Medical Center, Amsterdam, the Netherlands; Department of Informatics, Technische Universität Darmstadt, Darmstadt, Germany (A.M.); Department of Radiology, Utrecht University Medical Center, Utrecht, the Netherlands (T.L.); and Institute of Diagnostic and Interventional Radiology, Universitätsklinikum Frankfurt, Frankfurt, Germany (T.J.V., A.M.B.).
Radiographics. 2021 May-Jun;41(3):840-857. doi: 10.1148/rg.2021200151. Epub 2021 Apr 23.
Artificial intelligence techniques involving the use of artificial neural networks-that is, deep learning techniques-are expected to have a major effect on radiology. Some of the most exciting applications of deep learning in radiology make use of generative adversarial networks (GANs). GANs consist of two artificial neural networks that are jointly optimized but with opposing goals. One neural network, the generator, aims to synthesize images that cannot be distinguished from real images. The second neural network, the discriminator, aims to distinguish these synthetic images from real images. These deep learning models allow, among other applications, the synthesis of new images, acceleration of image acquisitions, reduction of imaging artifacts, efficient and accurate conversion between medical images acquired with different modalities, and identification of abnormalities depicted on images. The authors provide an introduction to GANs and adversarial deep learning methods. In addition, the different ways in which GANs can be used for image synthesis and image-to-image translation tasks, as well as the principles underlying conditional GANs and cycle-consistent GANs, are described. Illustrated examples of GAN applications in radiologic image analysis for different imaging modalities and different tasks are provided. The clinical potential of GANs, future clinical GAN applications, and potential pitfalls and caveats that radiologists should be aware of also are discussed in this review. RSNA, 2021.
人工智能技术涉及使用人工神经网络，也就是深度学习技术，预计将对放射学产生重大影响。深度学习在放射学中最令人兴奋的一些应用是利用生成对抗网络（GAN）实现的。GAN 由两个人工神经网络组成，它们共同优化，但目标相反。一个神经网络，即生成器，旨在合成无法与真实图像区分开的图像。第二个神经网络，即鉴别器，旨在区分这些合成图像和真实图像。这些深度学习模型除了其他应用外，还可以用于新图像的合成、图像采集的加速、成像伪影的减少、不同模态采集的医学图像之间的高效准确转换以及图像上异常的识别。作者介绍了 GAN 和对抗性深度学习方法。此外，还描述了 GAN 可用于图像合成和图像到图像转换任务的不同方式，以及条件 GAN 和循环一致 GAN 的基本原理。文中还提供了 GAN 在不同成像方式和不同任务的放射影像分析中的应用实例。本文还讨论了 GAN 的临床潜力、未来的临床 GAN 应用以及放射科医生应该注意的潜在陷阱和注意事项。RSNA，2021 年。
