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

Abstract Minimizing acquisition time and motion-artifacts remains challenging in magnetic resonance imaging (MRI) with demands on high-resolution images for accurate diagnosis and treatment. Deep learning-based image restoration offers promising solution by generating high-resolution and artifact-free MR images from low-resolution or motion-corrupted data. To facilitate practical deployment in clinical workflows, this study presents a time-/GPU-efficient framework using 2D network (TS-RCAN) for pseudo-3D MRI super-resolution reconstruction (SRR) and motion-artifact reduction (MAR). Optimal down-sampling factors were identified to balance SRR accuracy and acquisition time. MAR training used a standardized method to induce controllable motion-artifacts of varying severity. Network performance was benchmarked against state-of-the-art 3D networks. Results showed the down-sampling factor $$1\times 1\times 2$$ for $$\times 2$$ acceleration and $$2\times 2\times 2$$ for $$\times 4$$ acceleration achieved optimal SRR performance. TS-RCAN outperformed most 3D networks by > 0.01/1.5 dB in SSIM/PSNR while reducing GPU load and inference time by up to 90%. For MAR, TS-RCAN exceeded UNet by up to 0.014/1.48 dB in SSIM/PSNR. Additionally, uncertainty estimation correlated with image quality metrics, enabling accuracy prediction without ground truth. TS-RCAN provides an efficient, accurate framework for SRR and MAR with practical relevance to clinical MRI, and offers a flexible basis for future extension to other imaging contrasts and pathological cases.