Evaluating Procedures for Establishing Generative Adversarial Network-based Stochastic Image Models in Medical Imaging

TL;DR

This paper evaluates procedures for establishing GAN-based stochastic image models in medical imaging, focusing on realistic vessel simulation in angiography and highlighting the importance of medically relevant metrics.

Contribution

It introduces a framework for assessing GAN-based SIMs in medical imaging using canonical models and emphasizes the need for objective, medically relevant evaluation metrics.

Findings

GANs can replicate medically relevant vessel statistics

Classical and medical metrics may yield different GAN evaluation results

Objective metrics are crucial for assessing GAN fidelity in medical imaging

Abstract

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.