This paper proposes a deep learning approach for accelerating magnetic resonance imaging (MRI) using a large number of existing high quality MR images as the training datasets. An off-line convolutional neural network is designed and trained to identify the mapping relationship between the MR images obtained from zero-filled and fully-sampled k-space data. The network is not only capable of restoring fine structures and details but is also compatible with online constrained reconstruction methods. Experimental results on real MR data have shown encouraging performance of the proposed method for efficient and effective imaging.

Both parallel MRI and compressed sensing (CS) are emerging techniques to accelerate conventional MRI by reducing the number of acquired data. The combination of parallel MRI and CS for further acceleration is of great interest. In this paper, we propose a novel method to combine sensitivity encoding (SENSE), one of the standard methods for parallel MRI, and compressed sensing for rapid MR imaging (SparseMRI), a recently proposed method for applying CS in MR imaging with Cartesian trajectories. The proposed method, named CS-SENSE, sequentially reconstructs a set of aliased reducedfield-of-view images in each channel using SparseMRI and then reconstructs the final image from the aliased images using Cartesian SENSE. The results from simulations and phantom and in vivo experiments demonstrate that CS-SENSE can achieve a reduction factor higher than those achieved by SparseMRI and SENSE individually and outperform the existing method that combines parallel MRI and CS. Magn Reson Med 62:1574 -1584, 2009.

No abstract

This paper proposes a multi-channel image reconstruction method, named DeepcomplexMRI, to accelerate parallel MR imaging with residual complex convolutional neural network. Different from most existing works which rely on the utilization of the coil sensitivities or prior information of predefined transforms, DeepcomplexMRI takes advantage of the availability of a large number of existing multi-channel groudtruth images and uses them as target data to train the deep residual convolutional neural network offline. In particular, a complex convolutional network is proposed to take into account the correlation between the real and imaginary parts of MR images. In addition, the kspace data consistency is further enforced repeatedly in between layers of the network. The evaluations on in vivo datasets show that the proposed method has the capability to recover the desired multi-channel images. Its comparison with state-of-theart method also demonstrates that the proposed method can reconstruct the desired MR images more accurately.