Comparison of volume-selective z-shim and conventional EPI in fMRI studies using face stimuli

Abstract

Single-shot gradient recalled echo planar imaging (EPI) is the primary tool for functional magnetic resonance imaging (fMRI). The image often suffers from signal drop near the air-tissue interface, such as the amygdala and regions of the orbitofrontal lobe. An effective way to correct for this type of artifact is by applying multi-shot EPI using different z-shimming. Unfortunately, the scanning efficiency is significantly lowered. More recently, a new technique called volume-selective z-shim was proposed to implement z-shim compensation to only specific slices with large susceptibility effects. The high imaging efficiency of volume selective z-shim makes it possible to substitute conventional EPI for whole brain studies. In this study two fMRI experiments were conducted to compare volume- selective z-shim and conventional EPI while subjects performed tasks on face stimuli. The comparison was focused on three brain regions: amygdala, hippocampus, and fusiform gyrus. Our results indicate that despite fewer volumes collected during the same amount of scan time, volume-selective z-shim showed statistically higher activation in brain regions susceptible to signal loss while minimal differences were observed elsewhere.