ViGU: Vision GNN U-Net for Fast MRI

TL;DR

This paper introduces ViGU, a novel graph neural network-based U-Net for fast MRI that captures irregular anatomical features more effectively than traditional CNNs and Transformers, with improved performance and explainability.

Contribution

The paper presents a new Vision GNN U-Net architecture for MRI that outperforms existing CNN and GAN models and offers better interpretability and computational efficiency.

Findings

ViGU outperforms CNN and GAN-based methods in MRI reconstruction.

ViGU-GAN further improves image quality with adversarial training.

The graph structure enhances interpretability of feature extraction.

Abstract

Deep learning models have been widely applied for fast MRI. The majority of existing deep learning models, e.g., convolutional neural networks, work on data with Euclidean or regular grids structures. However, high-dimensional features extracted from MR data could be encapsulated in non-Euclidean manifolds. This disparity between the go-to assumption of existing models and data requirements limits the flexibility to capture irregular anatomical features in MR data. In this work, we introduce a novel Vision GNN type network for fast MRI called Vision GNN U-Net (ViGU). More precisely, the pixel array is first embedded into patches and then converted into a graph. Secondly, a U-shape network is developed using several graph blocks in symmetrical encoder and decoder paths. Moreover, we show that the proposed ViGU can also benefit from Generative Adversarial Networks yielding to its variant ViGU-GAN. We demonstrate, through numerical and visual experiments, that the proposed ViGU and GAN variant outperform existing CNN and GAN-based methods. Moreover, we show that the proposed network readily competes with approaches based on Transformers while requiring a fraction of the computational cost. More importantly, the graph structure of the network reveals how the network extracts features from MR images, providing intuitive explainability.