Various AI models are available today that are capable of reading medical data e.g. images and diagnosing the disease. However, these models can be difficult to train at times because the amount of available data is limited. We wanted to address this by implementing special AI models, called generative adversarial networks or GANs, which create synthetic data. These are fake data samples that do not exist in real life.
The benefit of this strategy is that they can be used to augment existing datasets and hopefully improve diagnostic performance. One of the challenges with such methods is how can we evaluate if a synthetic image looks good or not. This is what this quiz is all about. We are going to present to you an image, and it may be real or synthetic: you have to guess which is which! You may do this by selecting the option : Real / Unsure / Synthetic.