Fast dynamic magnetic resonance imaging using tagging RF pulses

A critical requirement for dynamic magnetic resonance imaging (MRI) is to reduce image acquisition times while maintaining high spatial resolutions to capture the underlying process with high-information rates. This paper presents a sparse signal recovery based fast MRI method which uses: 1) dictionary learning for sparse representation of signals for encoding of data redundancy in physiological functions and, 2) a tagging radio-frequency pulses based novel MR signal encoding formulation to uniformly sample the k-space, even at high acceleration factors. The preliminary results of dynamic MR image recovery experiments using tagging based MR signal acquisition method on an in-vivo myocardial perfusion dataset outperforms the equivalent dynamic MRI method implemented with variable density k-space under-sampling.