Summary:
- This article discusses a new type of artificial intelligence (AI) system called a "generative adversarial network" (GAN) that can be used to generate realistic-looking medical images, such as X-rays and MRI scans.
- The GAN system is trained on a large dataset of real medical images, and it can then generate new, synthetic images that are virtually indistinguishable from the real ones.
- This technology could be used to augment existing medical imaging datasets, which are often limited in size, and to create training data for AI-based medical image analysis algorithms.