A Modified Generative Adversarial Network Using Spatial and Channel-Wise Attention for CS-MRI Reconstruction

Abstract

Compressed sensing (CS) can speed up the magnetic resonance imaging (MRI) process and reconstruct high-quality images from under-sampled k-space data. However, traditional CS-MRI suffers from slow iterations and artifacts caused by noise when the acceleration factor is high. Currently, deep learning has been introduced to address these issues. Although some improvements have been achieved, the reconstruction problem under high under-sampling rates has not been solved. Thus, our study proposed a novel CS-MRI reconstruction method called RSCA-GAN. The generator of RSCA-GAN is a residual U-net consisting of both spatial and channel-wise attention. The proposed RSCA-GAN was compared to the zero-filling, DAGAN, RefineGAN, and RCA-GAN using both cartesian and non-cartesian undersampling masks on brain and knee datasets. The sampling rates of cartesian masks are 25%, 16.7%, and 12.5% and the sampling rates of spiral and radial masks are 30%, 20%, and 10%. Peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and normalized mean square error (NMSE) were used to evaluate the reconstructed image quality, and the rank-sum test was adopted to evaluate the significant difference among different approaches. P < 0.05 indicated statistical significance. The results demonstrated that RSCA-GAN outperforms the other approaches for all the quantitative metrics. Thus, RSCA-GAN exhibits excellent reconstruction performance at high under-sampling rates and is suitable for clinical applications.

Region-Guided Channel-Wise Attention Network for Accelerated MRI Reconstruction

“…To further demonstrate the role of RG-CAM, the variant, dubbed "w/o RG", adopts a single kernel to perform channel-wise attention without using the spatial guiding masks, similar to [27,17,13]. From Table 2, we found that the incorporation of the region-guided mechanism enhances the model performance, comparing "w/o RG" and "proposed".…”

“…Owing to its flexible design, RG-CAM can be easily combined with other network structures, e.g. the spatial attention modules [33,27,17], to retrieve further improvements.…”

“…reconstruction tasks. Alternatively, channel-wise attention models the inter-plays between feature channels efficiently and can potentially provide better MRI reconstructions [13,19,17], whereas the spatial information is not considered, limiting the representation capacity of attention over channels.…”

A Systematic Literature Review on Applications of GAN-Synthesized Images for Brain MRI

“…Traditional CS-MRI is affected by slow iterations and noise-induced artifacts during the high acceleration factor. The RSCA-GAN uses spatial and channel-wise attention with long skip connections to improve the quality at each stage, accelerating the reconstruction process and removing the artifacts brought by fast-paced under-sampling [118]. Parallel imaging integrated with the GAN model (PI-GAN) and transfer learning accelerates MRI imaging with under-sampling in the kspace.…”

Edge-enhanced dual discriminator generative adversarial network for fast MRI with parallel imaging using multi-view information

“…Yuan et al [ 46 ] incorporated the self-attention mechanism into the generator for a global understanding of images and improved the discriminator for the utilisation of prior knowledge. Li et al [ 47 ] proposed RSCA-GAN for fast MRI reconstruction, where both spatial and channel-wise attention mechanisms were applied in the generator. This team also applied the GAN-based model with attention mechanisms in the super-resolution task [ 48 ].…”