Research On Fast Magnetic Resonance Imaging Based On Deep Learning

Magnetic resonance imaging(MRI)with non-ionizing and non-radiating nature is capable of providing rich anatomical and functional information and is an indispensable tool for medical diagnosis,disease staging,and clinical research.However,the wide application of MRI in clinical practice is restricted by long scanning time.Thus,fast imaging has been constantly one of the emphases in the MRI technology.The existing multi-coil parallel imaging and partial k-space data reconstruction techniques decrease acquisition times by reducing the amount of phase encoding required.The parallel imaging utilizes different the spatial sensitivity of multiple coils.In the partial k-space data reconstruction techniques,diverse prior information as regularizations is incorporated into the reconstruction equation inside.These methods have achieved some success.However,serious aliasing artifacts may still occur in case of a high acceleration factor.Therefore,a means of accelerating the imaging rate while ensuring imaging accuracy should be devised.The extensive application of deep learning in recent years has demonstrated its powerful image processing capabilities.In particular,with the improvement of computer computing capabilities and the existing big data foundation,convolutional neural networks(CNN)excel in automatic feature extraction and nonlinear correlation description.The paper focuses on deep learning techniques for the speed and accuracy of fast magnetic resonance imaging.The research of parallel high-resolution magnetic resonance imaging based on convolutional neural network was carried out,and a fast MRI method based on CNN(CNN-MRI)was proposed.It can accelerate the reconstruction and ensure the accuracy of the image.The reconstruction results were quantitatively evaluated in terms of peak signal-to-noise ratio(PSNR),structural similarity(SSIM),and root-mean-square error(RMSE)to determine the performance of the proposed scheme.The main work and results of the thesis are as follows:(1)We designed a multi-coil CNN to exploit the local correlation in multi-channel images by using prior knowledge from numerous existing fully-sampled multi-coil data.The entire research contents include two main parts as follows:off-line training and online imaging.(2)Unlike the popular parallel imaging or compressed sensing technique,which exploits sensitivity and sparsity for fast MRI,CNN-MRI learns the end-to-end mapping between the MR images reconstructed from the undersampled and fully-sampled K-space data from huge offline acquisitions,and then aids in exacting online fast imaging with the learned mapping prior.Therefore,conventional sampling methods may not be the optimal undersampling trajectory for CNN-MRI,so we proposed a new trajectory scheme,namely,hamming filtered asymmetrical 1D partial Fourier sampling.(3)We compared the undersampling trajectory proposed with the traditional undersampling trajectory.Experimental results show that the proposed undersampling pattern performs better than the traditional sampling trajectory.The parameters of convolution neural network are optimized and compared,and we also shift the proposed undersampled trajectory to find the best undersampling pattern.In addition,we compared the proposed method to the classical parallel imaging methods,Synthetically,the mean quantitative and visual comparisons show that the proposed method produces superior quality with the least time,with a reconstruction rate of faster than five times.