Improved PARS-based PMRI image reconstruction as linear approximation

Document Type

Conference Proceeding

Publication Date

12-1-2012

Abstract

In this research, an improved approach to reconstruct high resolution brain images from Parallel Magnetic Resonance Imaging (pMRI) data is proposed based on the mathematical understanding that the PARS algorithms can be recast as linear approximates. It extends the PARS reconstruction method and is called IPARS in this paper. Our new improvements are in the following two aspects. First, localized significant coefficients of the Fourier series are selected and used for the reconstruction. Second, the maximum information from ACS lines have been retrieved and used for generating the coefficients for the reconstruction. The experiments have shown improved image reconstruction quality.

This document is currently not available here.

Share

COinS