Performance of a GPU- and Time-Efficient Pseudo 3D Network for Magnetic Resonance Image Super-Resolution and Motion Artifact Reduction

TL;DR

This paper introduces a GPU-efficient pseudo-3D deep learning network, TS-RCAN, for MRI super-resolution and motion artifact reduction, achieving high accuracy with reduced computational load and inference time, suitable for clinical use.

Contribution

The study presents a unified 2D deep learning framework with a novel TS-RCAN architecture that outperforms existing methods in MRI image enhancement tasks while being time- and GPU-efficient.

Findings

Optimal down-sampling factors identified for MRI acceleration.

TS-RCAN outperforms state-of-the-art 3D networks in SSIM and PSNR.

Significant reduction in GPU load and inference time achieved.

Abstract

Shortening acquisition time and reducing motion artifacts are the most critical challenges in magnetic resonance imaging (MRI). Deep learning-based image restoration has emerged as a promising solution capable of generating high-resolution and motion-artifact-free MRI images from low-resolution images acquired with shortened acquisition times or from motion-artifact-corrupted images. To facilitate clinical integration, a time- and GPU-efficient network with reliable accuracy is essential. In this study, we adopted a unified 2D deep learning framework for pseudo-3D MRI image super-resolution reconstruction (SRR) and motion artifact reduction (MAR). The optimal down-sampling factors to optimize the acquisition time in SRR were identified. Training for MAR was performed using publicly available in vivo data, employing a novel standardized method to induce motion artifacts of varying severity in a controlled way. The accuracy of the network was evaluated through a pixel-wise uncertainty map, and performance was benchmarked against state-of-the-art methods. The results demonstrated that the down-sampling factor of 1x1x2 for x2 acceleration and 2x2x2 for x4 acceleration was optimal. For SRR, the proposed TS-RCAN outperformed the 3D networks of mDCSRN and ReCNN, with an improvement of more than 0.01 in SSIM and 1.5 dB in PSNR while reducing GPU load by up to and inference time by up to 90%. For MAR, TS-RCAN exceeded UNet's performance by up to 0.014 in SSIM and 1.48 dB in PSNR. Additionally, TS-RCAN provided uncertainty information, which can be used to estimate the quality of the reconstructed images. TS-RCAN has potential use for SRR and MAR in the clinical setting.