1D Probabilistic Undersampling Pattern Optimization for MR Image Reconstruction
This work addresses MRI acceleration for clinical applications, but it is incremental as it builds on prior separate optimizations of sampling and reconstruction.
The paper tackles the problem of long scanning times and motion artifacts in 3D MRI by proposing a method to simultaneously optimize undersampling patterns and reconstruction models, resulting in improved quantitative and qualitative image recovery compared to existing strategies.
Magnetic resonance imaging (MRI) is mainly limited by long scanning time and vulnerable to human tissue motion artifacts, in 3D clinical scenarios. Thus, k-space undersampling is used to accelerate the acquisition of MRI while leading to visually poor MR images. Recently, some studies 1) use effective undersampling patterns, or 2) design deep neural networks to improve the quality of resulting images. However, they are considered as two separate optimization strategies. In this paper, we propose a cross-domain network for MR image reconstruction, in a retrospective data-driven manner, under limited sampling rates. Our method can simultaneously obtain the optimal undersampling pattern (in k-space) and the reconstruction model, which are customized to the type of training data, by using an end-to-end learning strategy. We propose a 1D probabilistic undersampling layer, to obtain the optimal undersampling pattern and its probability distribution in a differentiable way. We propose a 1D inverse Fourier transform layer, which connects the Fourier domain and the image domain during the forward pass and the backpropagation. In addition, by training 3D fully-sampled k-space data and MR images with the traditional Euclidean loss, we discover the universal relationship between the probability distribution of the optimal undersampling pattern and its corresponding sampling rate. Experiments show that the quantitative and qualitative results of recovered MR images by our 1D probabilistic undersampling pattern obviously outperform those of several existing sampling strategies.