Human brain high-resolution diffusion MRI with optimized slice-by-slice B0 field shimming in head-only high-performance gradient MRI systems

TL;DR

This work tackles $B_0$ inhomogeneity challenges in high-resolution diffusion MRI on head-only gradient systems by introducing brain tissue–selective, dynamic slice-by-slice shimming that accounts for the actual gradient fields. The method uses fast deep-learning brain segmentation to define the shimming ROI and compares linear-gradient assumptions against actual gradient-field models, evaluating performance with phantom and human diffusion MRI (PGSE and OGSE at 55 Hz) on a 3T head-only gradient system. Results show reductions in $Delta B_0$ and voxel displacement, with the greatest distortion mitigation achieved using two-shot EPI combined with dynamic shimming, and improvements in diffusion metrics in key brain regions such as the brainstem and cerebellum. The findings support the potential for improved accuracy of high-resolution diffusion MRI and pave the way for broader application to other MR modalities, although precise gradient-field mapping and higher-order effects remain areas for future work.

Abstract

The purpose of this study is to propose a brain tissue-selective, optimized slice-by-slice B0 field shimming for high-resolution brain diffusion MRI. We incorporated actual gradient fields of X, Y, and Z gradient coils in the calculation of the shimming coefficients in dynamic slice-by-slice B0 field shimming to minimize B0 field inhomogeneity (i.e., Delta B0) in deep-learning segmented brain tissues. Diffusion MRI with oscillating gradient spin echo (OGSE) at 55 Hz and pulsed gradient spin echo (PGSE) (approximated at 0 Hz) were obtained in phantoms and healthy volunteers using a head-only high-performance gradient 3T MRI system. In each diffusion MRI acquisition, standard static volumetric shimming and the proposed shimming method were applied separately, and mean/axial/radial diffusivities (MD/AD/RD) and fractional anisotropy (FA) were estimated. In phantom, the root-mean-square of Delta B0 in areas with high gradient nonlinearity was reduced by 7 Hz when incorporating actual gradient field in dynamic shimming. Compared to static shimming, dynamic shimming reduced root-mean-square of voxel displacement of each slice by a maximum of 5-10 voxels in single-shot EPI acquisition at 1-2 mm in-plane resolution in phantom, and a maximum of 3 voxels in human brains. Improved accuracy of MD/AD/RD/FA in the superior region of the brain, brainstem, and cerebellum were observed by applying dynamic shimming and/or two-shot EPI acquisition. MD(55 Hz)-MD(0 Hz) showed higher values in T2 FSE hypo-intensity region by applying dynamic shimming. We concluded that diffusion MRI with brain tissue-selective, dynamic slice-by-slice B0 effectively improves the accuracy of diffusivity characterization in high-resolution images.

Figures (7)

• Figure 1: Imaging setup in a head-only high-performance gradient MRI systems for phantoms (top) and human volunteers (bottom) using (A) static and (B) dynamic shimming. (C) Workflow of dynamic $B_0$ field shimming, where an axial $B_0$ field map is obtained, and the brain is segmented from the magnitude images using a deep-learning based algorithm. The resulting masked field map is used to calculate the slice-specific shimming coefficients, which are then applied during a second $B_0$ field mapping sequence and diffusion EPI sequences.
• Figure 2: $B_0$ field map analysis of a brain-shaped phantom. (A-F) Axial, coronal reformat, and sagittal reformat of the original $\Delta B_0$, predicted post-shim $\Delta B_0$ using linear/actual gradients, measured post-shim $\Delta B_0$ using linear/actual gradients, and the difference between the measured post-shim $\Delta B_0$ with linear and actual gradient fields. (G-I) Mean, standard deviation, and root-mean-square of $\Delta B_0$ in each slice. Voxel displacements were calculated using an echo spacing of 344 $\mu s$ and single-shot EPI acquisition in the presence of three $B_0$ maps, i.e., (J) the original $\Delta B_0$, (K) measured post-shim $\Delta B_0$ using linear gradients, (L) measured post-shim $\Delta B_0$ using actual gradients.
• Figure 3: $B_0$ field map analysis of a healthy volunteer's brain (i.e., Subject #1). (A-F) Axial, coronal reformat, and sagittal reformat of the original $\Delta B_0$, predicted post-shim $\Delta B_0$ using linear/actual gradients, measured post-shim $\Delta B_0$ using linear/actual gradients, and the difference between the measured post-shim $\Delta B_0$ with linear and actual gradient fields. (G-I) Mean, standard deviation, and root-mean-square of $\Delta B_0$ in each slice of brain tissues. Voxel displacements were calculated using an echo spacing of 344 $\mu s$ and single-shot EPI acquisition in the presence of three $B_0$ maps, i.e., (J) the original $\Delta B_0$, (K) measured post-shim $\Delta B_0$ using linear gradients, (L) measured post-shim $\Delta B_0$ using actual gradients.
• Figure 4: $B_0$ map and $T_2$-weighted EPI at 2 x 2 x 2 $mm^3$ resolution and 1 x 1 x 2 $mm^3$ resolution, and $T_2$ FSE of representative slices of the mini-ACR phantom in axial (A) and sagittal (B) views. The axial slice was noted by the red line in the sagittal $T_2$ FSE image. $T_2$-weighted EPI with dynamic shimming was acquired using the actual gradient field in the calculation of shimming coefficients. The blue lines indicate the maximum voxel displacement of the hollow rectangle compared to $T_2$ FSE, outlined in red. The voxel displacement in the 2 x 2 x 2 $mm^3$ EPI (C) and 1 x 1 x 2 $mm^3$ EPI (D) was calculated based on $\Delta B_0$ and EPI acquisition parameters in Table \ref{['table1']}.
• Figure 5: $B_0$ map, $T_2$-weighted EPI at 2 x 2 x 2 $mm^3$ resolution and 1 x 1 x 2 $mm^3$ resolution, and $T_2$ FSE of representative slices of a healthy volunteer's brain (i.e., Subject #2) in axial (A) and sagittal (B) views. The axial slice was noted by the green line in the sagittal $T_2$ FSE image. $T_2$-weighted EPI with dynamic shimming was acquired using the actual gradient field in the calculation of shimming coefficients. The voxel displacement in the 2 x 2 x 2 $mm^3$ EPI (C) and 1 x 1 x 2 $mm^3$ EPI (D) was calculated based on $\Delta B_0$ and EPI acquisition parameters in Table \ref{['table1']}.