Premium
Designing hyperbolic secant excitation pulses to reduce signal dropout in gradient‐echo echo‐planar imaging
Author(s) -
Wastling Stephen J.,
Barker Gareth J.
Publication year - 2015
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.25444
Subject(s) - signal (programming language) , sensitivity (control systems) , pulse (music) , nuclear magnetic resonance , physics , resting state fmri , spin echo , echo (communications protocol) , dropout (neural networks) , signal to noise ratio (imaging) , imaging phantom , acoustics , magnetic resonance imaging , optics , computer science , medicine , machine learning , detector , engineering , radiology , programming language , computer network , electronic engineering
Purpose To design hyperbolic secant (HS) excitation pulses to reduce signal dropout in the orbitofrontal and inferior temporal regions in gradient‐echo echo‐planar imaging (GE‐EPI) for functional MRI (fMRI) applications. Methods An algorithm based on Bloch simulations optimizes the HS pulse parameters needed to give the desired signal response across the range of susceptibility gradients observed in the human head (approximately ±250 μT·m −1 ). The impact of the HS pulse on the signal, temporal signal‐to‐noise ratio, blood oxygen level–dependent (BOLD) sensitivity, and ability to detect resting state BOLD signal changes was assessed in six healthy male volunteers at 3T. Results The optimized HS pulse (μ = 4.25, β = 3040 Hz, A 0 = 12.3 μT, Δf = 4598 Hz) had a near uniform signal response for through‐plane susceptibility gradients in the range ±250 μT·m −1 . Signal, temporal signal‐to‐noise ratio, BOLD sensitivity, and the detectability of resting state networks were all partially recovered in the orbitofrontal and inferior temporal regions; however, there were signal losses of up to 50% in regions of homogeneous field (and signal loss from in‐plane susceptibility gradients remained). Conclusion The HS pulse reduced signal dropout and could be used to acquire task and resting state fMRI data without loss of spatial coverage or temporal resolution. Magn Reson Med 74:661–672, 2015. © 2014 Wiley Periodicals, Inc.