Diffusion‐prepared 3D gradient spin‐echo sequence for improved oscillating gradient diffusion MRI
Dan Wu, Dapeng Liu, Yi‐Cheng Hsu, Haotian Li, Yi Sun, Qin Qin, Yi Zhang
Abstract
Purpose
Oscillating gradient (OG) enables the access of short diffusion times for time‐dependent diffusion MRI (dMRI); however, it poses several technical challenges for clinical use. This study proposes a 3D oscillating gradient–prepared gradient spin‐echo (OGprep‐GRASE) sequence to improve SNR and shorten acquisition time for OG dMRI on clinical scanners.
Methods
The 3D OGprep‐GRASE sequence consisted of global saturation, diffusion encoding, fat saturation, and GRASE readout modules. Multiplexed sensitivity‐encoding reconstruction was used to correct the phase errors between multiple shots. We compared the scan time and SNR of the proposed sequence and the conventional 2D‐EPI sequence for OG dMRI at 30‐90‐mm slice coverage. We also examined the time‐dependent diffusivity changes with OG dMRI acquired at frequencies of 50 Hz and 25 Hz and pulsed‐gradient dMRI at diffusion times of 30 ms and 60 ms.
Results
The OGprep‐GRASE sequence reduced the scan time by a factor of 1.38, and increased the SNR by 1.74‐2.27 times compared with 2D EPI for relatively thick slice coverage (60‐90 mm). The SNR gain led to improved diffusion‐tensor reconstruction in the multishot protocols. Image distortion in 2D‐EPI images was also reduced in GRASE images. Diffusivity measurements from the pulsed‐gradient dMRI and OG dMRI showed clear diffusion‐time dependency in the white matter and gray matter of the human brain, using both the GRASE and EPI sequences.
Conclusion
The 3D OGprep‐GRASE sequence improved scan time and SNR and reduced image distortion compared with the 2D multislice acquisition for OG dMRI on a 3T clinical system, which may facilitate the clinical translation of time‐dependent dMRI.