Segmented Echo Planar Imaging Improves Detection of Subcortical Functional Connectivity Networks in the Rat Brain

Stefano Tambalo,Angelo Bifone,Giulia Scuppa

doi:10.1038/s41598-018-37863-2

Abstract

Susceptibility artifacts in the vicinity of aural and nasal cavities result in significant signal drop-out and image distortion in echo planar imaging of the rat brain. These effects may limit the study of resting state functional connectivity in deep brain regions. Here, we explore the use of segmented EPI for resting state fMRI studies in the rat, and assess the relative merits of this method compared to single shot EPI. Sequences were evaluated in terms of signal-to-noise ratio, geometric distortions, data driven detection of resting state networks and group level correlations of time series. Multishot imaging provided improved SNR, temporal SNR and reduced geometric distortion in deep areas, while maintaining acceptable overall image quality in cortical regions. Resting state networks identified by independent component analysis were consistent across methods, but multishot EPI provided a more robust and accurate delineation of connectivity patterns involving deep regions typically affected by susceptibility artifacts. Importantly, segmented EPI showed reduced between-subject variability and stronger statistical significance of pairwise correlations at group level over the whole brain and in particular in subcortical regions. Multishot EPI may represent a valid alternative to snapshot methods in functional connectivity studies, particularly for the investigation of subcortical regions and deep gray matter nuclei.

Highlights

The phase difference of two images acquired at different echo times
We propose segmented EPI as a viable alternative to conventional echo planar imaging as it shows improved SNR, temporal SNR and a more robust and accurate delineation of deep brain regions, whose detection is normally negatively affected by B0 inhomogeneity
Preclinical applications of functional magnetic resonance imaging are the result of a well-balanced combinations of experimental, technical and ethical constraints

Summary

Introduction

The phase difference of two images acquired at different echo times. Each voxel in the map is an estimation of the phase - and position - shift caused by field inhomogeneity. Higher bandwidths, increased gradients strengths and shorter echo spacing, all lead to faster k-space filling, leaving less time for phase errors to accumulate These fine adjustments of the pulse program are constrained by hardware limitations and, especially for human studies, by fast gradient field switching that in some cases could lead to peripheral nerve stimulation. We apply a robust and minimally invasive protocol to assess the relative benefits of single- and multishot EPI for resting state fMRI in the rat brain at 7 Tesla, with particular attention to subcortical and ventral regions, where susceptibility artifacts are most prominent These circuits play a crucial role in many psychiatric disorders, as they are involved in the processing of goal-directed behavior, movement control, cognition and emotion. We propose segmented EPI as a viable alternative to conventional echo planar imaging as it shows improved SNR, temporal SNR and a more robust and accurate delineation of deep brain regions, whose detection is normally negatively affected by B0 inhomogeneity

Methods

Results

Conclusion