Optimisation of SOUP-GAN and CSR-GAN for High Resolution MR Images Reconstruction

TL;DR

This paper enhances MRI image reconstruction quality using modified GAN architectures, SOUP-GAN and CSR-GAN, with architectural improvements and training strategies, demonstrating superior performance in detail preservation and noise reduction.

Contribution

Introduces architectural modifications and training enhancements to GAN models for improved high-resolution MRI reconstruction.

Findings

CSR-GAN achieves higher PSNR and SSIM, better detail reconstruction.

SOUP-GAN produces less noisy images with good structural fidelity.

Enhanced GAN models improve MRI diagnostic image quality.

Abstract

Magnetic Resonance (MR) imaging is a diagnostic tool used in modern medicine; however, its output can be affected by motion artefacts and may be limited by equipment. This research focuses on MRI image quality enhancement using two efficient Generative Adversarial Networks (GANs) models: SOUP-GAN and CSR-GAN. In both models, meaningful architectural modifications were introduced. The generator and discriminator of each were further deepened by adding convolutional layers and were enhanced in filter sizes as well. The LeakyReLU activation function was used to improve gradient flow, and hyperparameter tuning strategies were applied, including a reduced learning rate and an optimal batch size. Moreover, spectral normalisation was proposed to address mode collapse and improve training stability. The experiment shows that CSR-GAN has better performance in reconstructing the image with higher frequency details and reducing noise compared to other methods, with an optimised PSNR of 34.6 and SSIM of 0.89. However, SOUP-GAN performed the best in terms of delivering less noisy images with good structures, achieving a PSNR of 34.4 and SSIM of 0.83. The obtained results indicate that the proposed enhanced GAN model can be a useful tool for MR image quality improvement for subsequent better disease diagnostics.