Abstract:Multi-Contrast Magnetic Resonance Imaging (MCMRI) utilizes the short-time reference image to facilitate the reconstruction of the long-time target one, providing a new solution for fast MRI. Although various methods have been proposed, they still have certain limitations. (1) Existing methods featuring the preset under-sampling patterns give rise to redundancy between multi-contrast images and limit their model performance. (2) Most methods focus on the information in the image domain, prior knowledge in the k-space domain has not been fully explored. (3) Most networks are manually designed and lack certain physical interpretability. To address these issues, we propose a joint optimization of the under-sampling pattern and a deep-unfolding dual-domain network for accelerating MCMRI. Firstly, to reduce the redundant information and sample more contrast-specific information, we propose a new framework to learn the optimal under-sampling pattern for MCMRI. Secondly, a dual-domain model is established to reconstruct the target image in both the image domain and the k-space frequency domain. The model in the image domain introduces a spatial transformation to explicitly model the inconsistent and unaligned structures of MCMRI. The model in the k-space learns prior knowledge from the frequency domain, enabling the model to capture more global information from the input images. Thirdly, we employ the proximal gradient algorithm to optimize the proposed model and then unfold the iterative results into a deep-unfolding network, called MC-DuDoN. We evaluate the proposed MC-DuDoN on MCMRI super-resolution and reconstruction tasks. Experimental results give credence to the superiority of the current model. In particular, since our approach explicitly models the inconsistent structures, it shows robustness on spatially misaligned MCMRI. In the reconstruction task, compared with conventional masks, the learned mask restores more realistic images, even under an ultra-high acceleration ratio (×30). Code is available at https://github.com/lpcccc-cv/MC-DuDoNet.

DD-SSUN: A Dual-domain Self-supervised Unrolled Network for MRI Reconstruction

DuDoINet: Dual-Domain Implicit Network for Multi-Modality MR Image Arbitrary-scale Super-Resolution

Dual-domain self-supervised learning for accelerated non-Cartesian MRI reconstruction

Self-supervised Learning for MRI Reconstruction with a Parallel Network Training Framework

MD-Recon-Net: A Parallel Dual-Domain Convolutional Neural Network for Compressed Sensing MRI

Multi-Modal Multi-Slice Cooperative Dual-Domain Cascaded De-Aliasing Network for MR Imaging Reconstruction

DuDoUniNeXt: Dual-domain unified hybrid model for single and multi-contrast undersampled MRI reconstruction

An Unrolled Neural Network for Accelerated Dynamic MRI Based on Second-Order Half-Quadratic Splitting Model

Self-Supervised MRI Reconstruction with Unrolled Diffusion Models

Clean self-supervised MRI reconstruction from noisy, sub-sampled training data with Robust SSDU

DuDoRNet: Learning a Dual-Domain Recurrent Network for Fast MRI Reconstruction With Deep T1 Prior

DFUSNN: zero-shot dual-domain fusion unsupervised neural network for parallel MRI reconstruction

DPFNet: Fast Reconstruction of Multi-coil MRI Based on Dual Domain Parallel Fusion Network

DDoS-UNet: Incorporating temporal information using Dynamic Dual-channel UNet for enhancing super-resolution of dynamic MRI

Dual-domain faster Fourier convolution based network for MR image reconstruction

Multi-Contrast MRI Reconstruction Via Information-growth Holistic Unfolding Network

Joint Under-Sampling Pattern and Dual-Domain Reconstruction for Accelerating Multi-Contrast MRI

DCT-net: Dual-domain cross-fusion transformer network for MRI reconstruction

Rethinking Dual-Domain Undersampled MRI reconstruction: domain-specific design from the perspective of the receptive field

Self-supervised Deep Unrolled Reconstruction Using Regularization by Denoising