Meeting Banner
Abstract #2653

A Random Projection Approach to Highly Efficient GRAPPA Reconstruction

Jingyuan Lyu1, Yuchou Chang2, Leslie Ying1

1Department of Biomedical Engineering, Department of Electrical Engineering, The State University of New York at Buffalo, Buffalo, NY, United States; 2Department of Electrical Engineering and Computer Science, University of Wisconsin-Milwaukee, Milwaukee, WI, United States

In GRAPPA, the computational time increases with the number of channels and the amount of ACS data. To address this issue, different from the existing approaches that compress the large number of physical channels to fewer virtual channels, we propose to use random projections to reduce the dimension of the problem in the calibration step. Experimental results show that randomly projecting the data onto a lower-dimensional subspace yields results comparable to those of traditional GRAPPA, but is computationally less expensive.

Abstract PDF Poster PDF

Keywords

accelerated accurate acquisition address amount approaches approximately array auto available becomes biomedical buffalo calibration channel channels chosen code coefficients coil combine commonly comp comparable complexity compress compressed comprised compromising computation computational computationally computer cost curves decreases define denotes desirable determined developed dimension dimensional dimensions distribution drawn efficient electrical element engineering equation equations errors especially estimation evaluated execution expense expensive experimental exploit exponential fashion fewer form full grant head highly important in vivo issue king least linearly maintain making manually matrices matrix measured memory much normalized onto paper performance physical preserved probability problem process project projected projecting projection projections property propose proposed prospective quality random randomly rapidly reconstruction reconstructions reduce reduced reduction represents respectively restricted retrospectively satisfy satisfying save saving science sensing serious simulate since slice solved space specifically squared squares step still subset subspace sufficiently suggesting system tackle targeting theoretically theory traditional trans trio unknown useful usually vector versus virtual written yields