Meeting Banner
Abstract #2601

Improved L1-SPIRiT Using Tensor-Based Sparsity Basis

Zhen Feng1, Feng Liu2, Stuart Crozier2, He Guo1

1Dalian University of Technology, Dalian, Liaoning, China; 2The University of Queensland, Brisbane, Queensland, Australia

In the sequential combination of parallel imaging (PI) and compressed sensing (CS) MRI, the CS procedure is conventionally performed on individual coils. In fact, the individual coil data are sensitivity-weighted maps of the whole MRI image, therefore signal overlapping exists between coil data. In this work, we propose a novel sparsity basis to improve CS reconstruction through the exploitation of the inter-coil spatial redundancies. In addition, by introducing a new filter that separates the measured and reconstructed data during L2-norm optimization, noise and errors can be minimized in the sequential PI-CS method. The brain image study showed the promise of the new PI-CS scheme.

Abstract PDF

Keywords

always amplification anal array attributions augmented avoid basis brain carried china clear clearly coil coils collect collected column combination combined compact conjugate convex core corner correlation criterion cross dataset decomposed decomposition dense descent developed dimensional disc domain dynamic eight either elements eliminated error errors exist exploit exploited feeding field filter filtered firstly form forth fully gradient illustrate implement importantly improve improved indicates individually inter intermediate intra introduced issues iterative kept limited major make many maps matrices matrix measured moor named natural noise nonlinear nonzero novel omitted operation orthogonal package parallel partial patterns performance procedure process projection propose proposed pulse pure reconstructed reconstruction recorded recovered reduction redundancies redundancy repetition residual respectively sampled sampling scheme secondly sensing separate sequential sets simultaneously singular slice software space sparse sparsity spirit stacked stage static step studied studies subsequent successfully superior superiority supporting technology tensor terms third transform transforms treat true typical verify visualization wasted wavelet widely