INTRODUCTION

Previous experimental report has shown that diffusion-weighted imaging (DWI) using multiplexed sensitivity encoding(MUSE), a multi-shot segmented EPI technique, can reliably produce high-resolution DWI of the brain [1]. With the interleaved k-space acquisition in each shot and parallel imaging reconstruction, MUSE can provide higher resolution, lower distortion, as well as a better SNR compared with conventional ssEPI. However, shot-to-shot phase variations caused by subject motion can severely degrade the image quality in segmented EPI, which is a major hurdle in pelvis DWI. Indeed, segmented EPI has not mostly been implemented for non-neural DWI [1-3]. Therefore, the purpose of this study was to compare MUSE and conventional DWI techniques in prostate MRI. Using quantitative comparison, we will validate whether MUSE technique could provide prostate DWI with less noise and distortions in comparison with conventional DWI or not, and this validating could provide the possibly leading to better lesion detectability in prostate.

METHODS

Five patients (mean age, 72.2 years) were enrolled in this study. Data were acquired using a 3T MR scanner (SIGNA Pioneer; GE Healthcare, Milwaukee, WI, USA) using a geometry-embracing method (GEM) 16-elements flexible array-coil and a 32-elements spine array-coil for signal detection, and a whole-body coil was used for radio-frequency excitation. MUSE-DWI sequence was acquired with the following parameters: TR/TE=7692/87.8 ms; FOV: 256 x 256 cm; matrix size: 128 x 128; slice thickness: 4 mm; total scan time: 4 min 30 sec. For quantitative comparing, the conventional ssEPI-DWI was acquired with the following parameters: TR/TE=11130/83.8 ms; FOV: 256 x 256 cm; matrix size: 118 x 118; slice thickness: 4 mm; total scan time: 4 min 26 sec. Three b-values (0, 800, and 1500) was acquired by those two DWI sequences. For spatial segmentation, high-resolution T2-weighted propeller images was acquired with the following parameters: TR/TE=7289/97.3 ms; FOV: 256 x 256; 320 x 320; slice thickness: 4 mm. All data processing steps were performed using MATLAB software. First, all the images were co-registered to eliminate the motion issue. Second, the manual spatial segmentation was applied in the high-resolution T2-weighted propeller images. Third, the segmented volume of prostate was applied into the B-null images by MUSE-DWI and conventional ss-DWI for signal-to-noise ratio (SNR) and distortion detection based on the overlapping between the segmented volume and the prostate regions by DWI images.

RESULTS and DISCUSSION

It’s obviously that images by MUSE-DWI showed less distortion than that by conventional ssEPI-DWI (Figure 1). In the prostate diffusion-weighted images with zero b-value, the images by MUSE-DWI showed significant less (p<0.023; paired t-test) distortion than that by conventional ssEPI-DWI (Figure 2). Agreed with our hypothesis, images by MUSE-DWI showed significant higher SNR than the images by conventional ssEPI-DWI (p<0.049; paired t-test; demonstrated on the Table1). MUSE-DWI can provide multi-shot prostate images with fewer distortions and higher SNR, and it could provide better image quality than conventional DWI images.

CONCLUSION

Compared with diffusion images in prostate by conventional ssEPI-DWI, MUSE-DWI can provide not only better spatial resolution but also better image quality that showed less distortions and higher SNR. Despite the relative longer acquisition time, MUSE-DWI could be useful for the detection and characterization of prostate lesion.

References

1. Chen N-K, Guidon A, Chang H-C, Song AW. A robust multi-shot scan strategy for high-resolution diffusion weighted MRI enabled by multiplexed sensitivity-encoding (MUSE). Neuroimage. 2013;72:41–7. 2. Chen X, Zhang Y, Cao Y, Sun R, Huang P, Xu Y, et al. A feasible study on using multiplexed sensitivity-encoding to reduce geometric distortion in diffusion-weighted echo planar imaging. Magn Reson Imaging. 2018;54:153–9. 3. Herbst M, Deng W, Ernst T, Stenger VA. Segmented simultaneous multi-slice diffusion weighted imaging with generalized trajectories. Magn Reson Med. 2017;78:1476–81.