Human liver CEST imaging at 7 T: Impact of B 1 + shimming

To explore the feasibility of CEST imaging in the human liver at 7 T with $B_1^{+}$ shimming.

CEST MRI was performed on a 7 T whole-body scanner with a parallel transmission (pTx) system in five healthy volunteers. Static pTx ( $B_1^{+}$ shimming) was applied to locally maximize the $B_1^{+}$ magnitude per input power within a given region of interest (ROI) of approximately 30 mm diameter (ROI_shim). Relaxation-compensated inverse magnetization transfer ratio (MTR_Rex) values were quantified for amide protons, guanidino protons, and relayed nuclear Overhauser effect signals based on five-pool Lorentzian fit analysis. MTR_Rex values were corrected for B₁ inhomogeneities using an absolute, accurate MR fingerprinting-based $B_1^{+}$ mapping technique.

Within the ROI_shim, reliable MTR_Rex values could be calculated for an average of 85% of voxels. The mean MTR_Rex values and corresponding coefficient of variations across the group are: 0.113 ± 0.009, 8.8% for amide; 0.167 ± 0.010, 6.3% for nuclear Overhauser effect; and 0.079 ± 0.010, 12.9% for guanidino. MTR_Rex values exhibit low variation between subjects, as reflected by low coefficient of variations.

In this study, we have demonstrated for the first time the feasibility of acquiring and quantifying relaxation-compensated CEST contrasts in the human liver at ultrahigh field. The application of static pTx effectively eliminates $B_1^{+}$ dropouts and allows for accurate CEST contrast quantification within the selected ROI. In addition, the proposed $B_1^{+}$ mapping technique shows efficacy for enhanced MTR_Rex B₁ corrections in the abdomen.