Temporal Signal-to-Noise Changes in Combined Multislice- and In-Plane-Accelerated Echo-Planar Imaging with a 20- and 64-Channel Coil

Abstract

Echo-planar imaging (EPI) is the most common method of functional MRI for acquiring the blood oxygenation level-dependent (BOLD) contrast, allowing the acquisition of an entire brain volume within seconds. However, because imaging protocols are limited by hardware (e.g., fast gradient switching), researchers must compromise between spatial resolution, temporal resolution, or whole-brain coverage. Earlier attempts to circumvent this problem included developing protocols in which slices of a volume were acquired faster (i.e., in-plane acceleration (S)) or simultaneously (i.e., multislice acceleration (M)). However, applying acceleration methods can lead to a reduction in the temporal signal-to-noise ratio (tSNR): a critical measure of signal stability over time. Using a 20- and 64-channel receiver coil, we show that enabling S-acceleration consistently yielded a substantial decrease in tSNR, regardless of the receiver coil, whereas M-acceleration yielded less pronounced tSNR decrease. Moreover, tSNR losses tended to occur in temporal, insular, and medial brain regions and were more noticeable with the 20-channel coil, while with the 64-channel coil, the tSNR in lateral frontoparietal regions remained relatively stable up to six-fold M-acceleration producing comparable tSNR to that of no acceleration. Such methodological explorations can guide researchers and clinicians in optimizing imaging protocols depending on the brain regions under investigation.

Figure 1

Changes in tSNR as a function of combined acceleration methods in a central region of interest (ROI) of the phantom. The x-axis represents the total acceleration factor (i.e., the product of M- and S-acceleration). Employing slice-acceleration leads to an immediate decrease in tSNR, whereas M-acceleration tends to yield decreases in tSNR only with higher acceleration factors. Moreover, the 64-channel coil appears to yield less tSNR decreases compared to the 20-channel coil.

Figure 2

Group-level changes in tSNR from the whole-brain average analysis. (A) Results plotted as a function of total acceleration, as in Fig. 1. (B) Results plotted as a function of the experimental conditions (i.e., not multiplying the M- and S-acceleration factors). Overall, the results reflect those from the phantom analysis, namely that enabling S-acceleration leads to more substantial decreases in tSNR; however, the difference between the two coils is not as pronounced in this analysis. Error bars represent one standard error of the mean.

Figure 3

Temporal SNR decreases at each of four thresholds (f.l.t.r. p < 1/31, 3/31, 6/31, 15/31, i.e., approx. 0.03, 0.1, 0.19, and 0.48) plotted (on a pseudo-log scale) as the percentage of voxels whose t-scores from each ΔtSNR map surpassed the FWER-corrected threshold for detecting tSNR decreases. Note that even at the more liberal FWER-corrected thresholds, the 64-channel coil with no S-acceleration yielded the least number of voxels with tSNR decreases, especially at intermediate M factors.

Figure 4

Uncorrected t-score maps (t₄ = ±3, p < 0.02) of tSNR changes presented for each combination of acceleration factors compared to the baseline S1-M1 (i.e., non-accelerated) projected onto a standard cortical surface for both the 20-channel coil (left) and 64-channel coil (right). Cold colors indicate a decrease in tSNR, while warm colors indicate an increase in tSNR. As predicted by the previous analyses, S-acceleration leads to the greatest—and most widespread—decreases in tSNR, though to a lesser extent when using the 64-channel coil. In all cases where tSNR does decrease, the temporal, insular, and medial regions appear most susceptible, in contrast to frontal and parietal regions.

Figure 5

Probability density functions of the group-averaged tSNR values resulting from different combinations of S- and M-acceleration factors for the 20-channel (blue) and 64-channel (red) coil.