Meeting Banner
Abstract #3966

Implications of MRSI Compressed Sensing Reconstruction on Spatial Resolution Using a Modified Technique

Keith Wachowicz1, 2, Amr Heikal3, B. Gino Fallone, 24

1Oncology, University of Alberta, Edmonton, Alberta, Canada; 2Medical Physics, Cross Cancer Institute, Edmonton, Alberta, Canada; 3Physics, University of Alberta, Edmonton, Alberta, Canada; 4Physics and Oncology, University of Alberta, Edmonton, Alberta, Canada

The implications of compressed sensing (CS) reconstruction in spectroscopic imaging on spatial resolution have not been quantitatively investigated in the literature. Modulation transfer analyis in this work is used to demonstrate that dramatic loss of modulation transfer across the range of spatial frequencies can occur with CS. A modification to the CS algorithm (CMaCS) involving further constraints on the randomization of k-space and mapping of unpaired conjugate data is introduced and shown to recover much of this lost modulation.

Keywords

acetate acquisition allowed allows although amount amplitude appearance array better black blue cancer choline clear community complete compressed conjugate constraint correction corrections cost cylinder degree derived diameter discontinuities displayed distribution dramatic encoding equivalent evaluated exhibit fact field firstly fluid fold frequencies frequency frequent function fundamental generated great greatly grid handling harmonic house illustrate imitate impact implementation implementations implemented implications independently input inside instability insufficient intensity introduce iterative limiting linear locations magenta magnitude mapping maps materials matrix measured medical modification modified modulation moreover much near nearly normalized object oncology operation outline output peak phantom phys physics pixel plot potential press prior probability proceeding qualitative quantified quantifies quantitative quantitatively radiology randomization randomized reconstruct reconstruction record recorded reduction replicas reports represent resolution response reversed robustness sampled sampling scenarios secondly sensing sinusoidally solution space sparsity spatial spectra spectral spectroscopic square structure suffer systems technical transfer transformed uniform unlikely unpaired variation varying voids wave years zero