Abstract:Background: Compressed sensing has been employed to accelerate magnetic resonance imaging by sampling fewer measurements. However, conventional iterative optimization-based CS-MRI are time-consuming for iterative calculations and often share poor generalization ability on multicontrast datasets. Most deep-learning-based CS-MRI focus on learning an end-to-end mapping while ignoring some prior knowledge existed in MR images. Purpose: We propose an iterative fusion model to integrate the image and gradient-based priors into reconstruction via convolutional neural network models while maintaining high quality and preserving the detailed information as well. Methods: We propose a gradient-enhanced fusion network (GFN) for fast and accurate MRI reconstruction, in which dense blocks with dilated convolution and dense residual learnings are used to capture abundant features with fewer parameters. Meanwhile, decomposed gradient maps containing obvious structural information are introduced into the network to enhance the reconstructed images. Besides this, gradient-based priors along directions X and Y are exploited to learn adaptive tight frames for reconstructing the desired image contrast and edges by respective gradient fusion networks. After that, both image and gradient priors are fused in the proposed optimization model, in which we employ the l2 -norm to promote the sparsity of gradient priors. The proposed fusion model can effectively help to capture edge structures in the gradient images and to preserve more detailed information of MR images. Results: Experimental results demonstrate that the proposed method outperforms several CS-MRI methods in terms of peak signal-to-noise (PSNR), the structural similarity index (SSIM), and visualizations on three sampling masks with different rates. Noteworthy, to evaluate the generalization ability, the proposed model conducts cross-center training and testing experiments for all three modalities and shares more stable performance compared than other approaches. In addition, the proposed fusion model is applied to other comparable deep learning methods. The quantitative results show that the reconstruction results of these methods are obviously improved. Conclusions: The gradient-based priors reconstructed from GFNs can effectively enhance the edges and details of under-sampled data. The proposed fusion model integrates image and gradient priors using l2 -norm can effectively improve the generalization ability on multicontrast datasets reconstruction.

Gradient-based Compressive Image Fusion

Image fusion in compressed sensing

Curvelet Based Hyperspectral Image Fusion

PCNN-Based Image Fusion in Compressed Domain

Ultrasonic image fusion using compressed sensing

Salience Preserving Image Fusion with Dynamic Range Compression.

Classification-based image-fusion framework for compressive imaging

Region-Based Image-Fusion Framework for Compressive Imaging

A multi-source image fusion algorithm based on gradient regularized convolution sparse representation

Research on GF-1 Remote Sensing IHS Image Fusion Algorithm Based on Compressed Sensing

EgeFusion: Towards Edge Gradient Enhancement in Infrared and Visible Image Fusion With Multi-Scale Transform

SCGRFuse: An infrared and visible image fusion network based on spatial/channel attention mechanism and gradient aggregation residual dense blocks

Compressive Image Fusion Based on Particle Swarm Optimization Algorithm

Fusion of Remote Sensing Image with Compressed Sensing Based on Wavelet Sparse Basis

Images Fusion Based on Block Compressed Sensing and Multiwavelet Transform

Fusion Research of Remote Sensing Image Based on Compressive Sensing

Deep compressed sensing MRI via a gradient-enhanced fusion model

A Variational Method for Multisource Remote-Sensing Image Fusion

Compressive fusion for remote-sensing images by block compressed sensing

MGFuse: An Infrared and Visible Image Fusion Algorithm Based on Multiscale Decomposition Optimization and Gradient-Weighted Local Energy

Salience Preserving Multifocus Image Fusion With Dynamic Range Compression