Evaluating Procedures for Establishing Generative Adversarial Network-based Stochastic Image Models in Medical Imaging
This work addresses the need for better evaluation metrics for GANs in medical imaging, which is incremental as it highlights issues without proposing new methods.
The paper tackled the problem of evaluating generative adversarial networks (GANs) for stochastic image models in medical imaging, specifically using angiography vessel simulations, and found that classical and medically relevant metrics can lead to different conclusions about GAN fidelity.
Modern generative models, such as generative adversarial networks (GANs), hold tremendous promise for several areas of medical imaging, such as unconditional medical image synthesis, image restoration, reconstruction and translation, and optimization of imaging systems. However, procedures for establishing stochastic image models (SIMs) using GANs remain generic and do not address specific issues relevant to medical imaging. In this work, canonical SIMs that simulate realistic vessels in angiography images are employed to evaluate procedures for establishing SIMs using GANs. The GAN-based SIM is compared to the canonical SIM based on its ability to reproduce those statistics that are meaningful to the particular medically realistic SIM considered. It is shown that evaluating GANs using classical metrics and medically relevant metrics may lead to different conclusions about the fidelity of the trained GANs. This work highlights the need for the development of objective metrics for evaluating GANs.