It is challenging and inspiring for us to achieve high spatiotemporal resolutions in dynamic cardiac magnetic resonance imaging (MRI). In this paper, we introduce two novel models and algorithms to reconstruct dynamic cardiac MRI data from under-sampledk-tspace data. In contrast to classical low-rank and sparse model, we use rank-one and transformed sparse model to exploit the correlations in the dataset. In addition, we propose projected alternative direction method (PADM) and alternative hard thresholding method (AHTM) to solve our proposed models. Numerical experiments of cardiac perfusion and cardiac cine MRI data demonstrate improvement in performance.
Accelerated cardiac cine MRI using locally low rank and finite difference constraints