Pyramid Convolutional RNN for MRI Reconstruction

Fast and accurate MRI image reconstruction from undersampled data is critically important in clinical practice. Compressed sensing based methods are widely used in image reconstruction but the speed is slow due to the iterative algorithms. Deep learning based methods have shown promising advances in recent years. However, recovering the fine details from highly undersampled data is still challenging. In this paper, we introduce a novel deep learning-based method, Pyramid Convolutional RNN (PC-RNN), to reconstruct the image from multiple scales. We evaluated our model on the fastMRI dataset and the results show that the proposed model achieves significant improvements than other methods and can recover more fine details.