Enhanced Magnetic Resonance Image Synthesis with Contrast-Aware Generative Adversarial Networks

TL;DR

This paper introduces a contrast-aware GAN that synthesizes realistic MR knee images conditioned on acquisition parameters, enabling adjustable contrast and supporting clinical, training, and AI data augmentation applications.

Contribution

The study presents a novel GAN model capable of generating MR images with variable contrast based on acquisition parameters, addressing limitations of prior models trained on fixed settings.

Findings

Experts mislabeled 40.5% of real and synthetic images, indicating high realism.

The model enables adjustable contrast synthesis for diverse clinical scenarios.

Synthetic images are comparable in quality to real images in visual tests.

Abstract

A Magnetic Resonance Imaging (MRI) exam typically consists of the acquisition of multiple MR pulse sequences, which are required for a reliable diagnosis. Each sequence can be parameterized through multiple acquisition parameters affecting MR image contrast, signal-to-noise ratio, resolution, or scan time. With the rise of generative deep learning models, approaches for the synthesis of MR images are developed to either synthesize additional MR contrasts, generate synthetic data, or augment existing data for AI training. However, current generative approaches for the synthesis of MR images are only trained on images with a specific set of acquisition parameter values, limiting the clinical value of these methods as various sets of acquisition parameter settings are used in clinical practice. Therefore, we trained a generative adversarial network (GAN) to generate synthetic MR knee images conditioned on various acquisition parameters (repetition time, echo time, image orientation). This approach enables us to synthesize MR images with adjustable image contrast. In a visual Turing test, two experts mislabeled 40.5% of real and synthetic MR images, demonstrating that the image quality of the generated synthetic and real MR images is comparable. This work can support radiologists and technologists during the parameterization of MR sequences by previewing the yielded MR contrast, can serve as a valuable tool for radiology training, and can be used for customized data generation to support AI training.