Resting state BOLD functional connectivity at 3T: spin echo versus gradient echo EPI.

Piero Chiacchiaretta,Antonio Ferretti,Emmanuel Andreas Stamatakis

doi:10.1371/journal.pone.0120398

Piero Chiacchiaretta, Antonio Ferretti + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0120398

Copy DOI

Abstract

Previous evidence showed that, due to refocusing of static dephasing effects around large vessels, spin-echo (SE) BOLD signals offer an increased linearity and promptness with respect to gradient-echo (GE) acquisition, even at low field. These characteristics suggest that, despite the reduced sensitivity, SE fMRI might also provide a potential benefit when investigating spontaneous fluctuations of brain activity. However, there are no reports on the application of spin-echo fMRI for connectivity studies at low field. In this study we compared resting state functional connectivity as measured with GE and SE EPI sequences at 3T. Main results showed that, within subject, the GE sensitivity is overall larger with respect to that of SE, but to a less extent than previously reported for activation studies. Noteworthy, the reduced sensitivity of SE was counterbalanced by a reduced inter-subject variability, resulting in comparable group statistical connectivity maps for the two sequences. Furthermore, the SE method performed better in the ventral portion of the default mode network, a region affected by signal dropout in standard GE acquisition. Future studies should clarify if these features of the SE BOLD signal can be beneficial to distinguish subtle variations of functional connectivity across different populations and/or treatments when vascular confounds or regions affected by signal dropout can be a critical issue.

Highlights

Since its discovery [1,2,3], functional magnetic resonance imaging based on blood oxygenation level-dependent (BOLD) contrast has been extensively used for mapping brain function in humans
After the first observation that spontaneous BOLD fluctuations in the left and right motor cortex are correlated in the absence of a task [4], resting-state functional magnetic resonance imaging (fMRI) has witnessed an exponential growth of interest
SE BOLD signals were replicable in response to replicated neuronal activities even at inter-stimulus intervals as short as ~1s, with no significant reduction in amplitude or increase in latency [26]. These findings suggested that the increased linearity and promptness of the SE BOLD signal could have potential benefits for quantitative assessment of the neurovascular coupling relationship and for connectivity studies [21,26]

Summary

Introduction

Since its discovery [1,2,3], functional magnetic resonance imaging (fMRI) based on blood oxygenation level-dependent (BOLD) contrast has been extensively used for mapping brain function in humans. The attractiveness of resting-state fMRI to study brain functional connectivity stems from the potential use as a biomarker for various diseases and from the easy of implementation even for problematic patient populations, since no task is required [5,6,7]. The BOLD signal is usually measured using gradient-echo (GE) T2Ã-weighted images due to their large sensitivity to deoxyhaemoglobin variations associated with the hemodynamic. MRI signal in spin-echo (SE) T2-weighted images is affected by the local deoxyhaemoglobin content but to a less extent [8,9,10], with a significant drop in sensitivity (about a factor of 3 at 3T) with respect to its GE counterpart [11,12,13]

Methods

Results

Conclusion