Premium
Improved combination of spiral‐in/out images for BOLD fMRI
Author(s) -
Glover Gary H.,
Thomason Moriah E.
Publication year - 2004
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.20016
Subject(s) - spiral (railway) , signal (programming language) , contrast (vision) , weighting , nuclear magnetic resonance , artificial intelligence , pattern recognition (psychology) , computer science , physics , neuroscience , mathematics , psychology , acoustics , mathematical analysis , programming language
Acquisitions with the spiral‐in/out technique result in two separate image timeseries obtained during the spiral‐in and spiral‐out trajectory. In uniform brain regions the two components have comparable signal and BOLD contrast and can be averaged, but in regions compromised by susceptibility effects where both signal and noise can differ in the two images other combination methods may be more effective. Here, several weighting schemes are compared for signal and activation contrast recovery in whole brain and prefrontal cortex using verbal working memory (seven subjects) and breathholding tasks (six subjects) scanned at 3 T. It was found that a statistically weighted combination based on activation maps derived separately from the spiral‐in and spiral‐out images provides activation volumes with increases of 33–59% over second‐choice signal‐weighted combination and 100–200% increases over spiral‐out acquisition alone, and that simple averaging is inferior to signal‐weighted combination. Magn Reson Med 51:863–868, 2004. © 2004 Wiley‐Liss, Inc.