Snapshot gradient-recalled echo-planar images of rat brains at long echo time at 9.4 T

Abstract

With improved B 0 homogeneity along with satisfactory gradient performance at high magnetic fields, snapshot gradient-recalled echo-planar imaging (GRE-EPI) would perform at long echo times (TEs) on the order of T2*, which intrinsically allows obtaining strongly T2*-weighted images with embedded substantial anatomical details in ultrashort time. The aim of this study was to investigate the feasibility and quality of long TE snapshot GRE-EPI images of rat brain at 9.4 T. When compensating for B 0 inhomogeneities, especially second-order shim terms, a 200 x 200 microm2 in-plane resolution image was reproducibly obtained at long TE (>25 ms). The resulting coronal images at 30 ms had diminished geometric distortions and, thus, embedded substantial anatomical details. Concurrently with the very consistent stability, such GRE-EPI images should permit to resolve functional data not only with high specificity but also with substantial anatomical details, therefore allowing coregistration of the acquired functional data on the same image data set.

Fig. 1

Comparison of the snapshot GRE-EPI images acquired with both first- and second-order shim terms (the bottom row, selected from Panel A) and those with first-order shim terms only (the top row) on the identical slices from the same animal. In this particular study, the second-order shim terms, Z²=−0.49×10⁻² mT/cm², YZ=0.02×10⁻² mT/cm², XZ=−1.09×10⁻² mT/cm², XY=−0.01×10⁻² mT/cm² and X²Y²=0.11×10⁻² mT/cm², were applied before acquiring the images in the bottom row.

Fig. 2

Direct comparison between snapshot GRE-EPI, GE images and FSE images from one rat brain. All images were acquired after second-order shim adjustments with identical slice thickness of 1 mm and 0.5 mm gap. GRE-EPI and GE images were acquired using 200×200 μm² in-plane resolutions, 30 ms TE and 2 s repetition time, but 30° flip angle and 2.5 ms acquisition time (per k-space line in read dimension) were used in GE and 90° flip angle and 0.3 ms acquisition time were used in GRE-EPI. The GRE-EPI images presented strong T₂* contrast yet minor susceptibility artifacts, consistent with that present in GE images.

Fig. 3

Multislice series of coronal (A) and transverse (B) GRE-EPI images acquired with TE>20 ms at 9.4 T. In Panel A, ten 0.5-mm-thick slices without gap (from bregma +0.5 mm to bregma −3.8 mm) were acquired with 200×200 μm² in-plane resolution, TR/TE=2000/30 ms, as described in Section 2. In Panel B, ten 0.5-mm-thick slices with 312×312 μm² in-plane resolution images (from bregma −1.3 mm to bregma −6.8 mm) were acquired with the similar parameters as in Panel A but with FOV=40 mm² and TE=27 ms. Both images with long TE presented with less ghosting artifact and geometric distortions, which were further evidenced by being overlaid with the corresponding structures identified from the rat brain atlas (red lines in panels). Fine structures that were readily identified include 3V, CA3, cc, cg, CPu, D3V, DG, ec, ic, LV, MnPO, TS and vhc (see details in the text), indicated with red arrows in Panel A.

Fig. 4

Temporal stability of snapshot GRE-EPI images. The ROIs, in which the signals were measured, are marked in boxes on those 200×200 μm² in-plane resolution and 1-mm slice thickness EPI images (A and B), which were acquired with two dummy scans and other parameters as described in Section 2. The corresponding percent deviations from the mean intensity from particular ROIs are shown as open squares for the left ROI and open diamonds for the right ROI. Filled circles denote both, respectively, on the side of each image. Over 10 min acquisition of 300 scans (TR=2 s), the percent deviations are mostly within 1.5% for two individual 16-pixel ROIs in Image A. However, the deviations of both 16-pixel ROIs (the 32-pixel ROI) in Image A, two individual 45-pixel ROIs and both 45-pixel ROIs (the 90-pixel ROI) in Image B mostly fall in 1%.

Fig. 5

Correlations between individual RMSD of each ROI with the corresponding volume from Fig. 4. A strong linear correlation (R=.92) was found with Y=3.0X+0.3, where Y is the RMSD of the percent deviations and X is $1/\sqrt{\text{pixels}}$, with “pixels” indicating the number of pixels of the corresponding ROI.