Parallel magnetic resonance imaging: characterization and comparison

Date

2005-11-01

Journal Title

Journal ISSN

Volume Title

Publisher

Texas A&M University

Abstract

Magnetic Resonance Imaging (MRI) is now increasingly being used for fast imaging applications such as real-time cardiac imaging, functional brain imaging, contrast enhanced MRI, etc. Imaging speed in MRI is mainly limited by different imaging parameters selected by the pulse sequences, the subject being imaged and the RF hardware system in operation. New pulse sequences have been developed in order to decrease the imaging time by a faster k-space scan. However, they may not be fast enough to facilitate imaging in real time. Parallel MRI (pMRI), a technique initially used for improving image SNR, has emerged as an effective complementary approach to reduce image scan-time. Five methods, viz., SENSE [Pruesmann, 1999], PILS [Griswold, 2000], SMASH [Sodickson, 1997], GRAPPA [Griswold, 2002] and SPACE RIP [Kyriakos, 2000]; developed in the past decade have been studied, simulated and compared in this research. Because of the dependence of the parallel imaging methods on numerous factors such as receiver coil configuration, k-space subsampling factor, k-space coverage in the imaging environment, there is a critical need to find the method giving the best results under certain imaging conditions. The tools developed in this research help the selection of the optimal method for parallel imaging depending on a particular imaging environment and scanning parameters. Simulations on real MR phased-array data show that SENSE and GRAPPA provide better image reconstructions when compared to the remaining techniques.

Description

Citation