Introduction

Sodium (²³Na) MRI can reveal valuable metabolic information¹ to complement structural proton MRI or MR fingerprinting (MRF). A technique to simultaneously acquire ²³Na MRI and ¹H MRF data was demonstrated recently^{2, 3}. In this work, we assessed the repeatability of proton density (PD), T₁, T₂ maps simultaneously acquired with a normalized sodium density image using 3D ¹H MRF/²³Na MRI3 in the brain at 7T.

Methods

Seven healthy volunteers (3 men, 4 women, mean age=35.7±8.2 years) were scanned three times in two sessions within a week. In the first session, the volunteers were scanned twice, with a short break between the scans. In the second session, the volunteers were scanned once.

The simultaneous 3D ¹H MRF/²³Na MRI sequence³ is based on a stack-of-stars sampling scheme⁴. Each TR (7.5ms), a non-selective ¹H pulse is applied, followed by one simultaneous readout for both nuclei (Fig.1). The images from both nuclei have the same slice thickness. The frequency encoding gradient moments are distributed such that we acquire a full radial trajectory for ¹H while acquiring a center-out readout for ²³Na, which allows for a factor ~2 between the ²³Na and ¹H resolutions since the ratio of their gyromagnetic ratios is ~4.

The proton flip angle (FA) train starts with an adiabatic inversion, followed by 275 non-selective excitations of variable FA (Fig. 1). The first segment (TR 1-100) includes FA up to 20°. The second segment (TR 101-200) contains FA up to 90°. A 50-TR delay is inserted to recover magnetization before a 3rd segment (TR 251 ~ 325), with FA train up to 60°. A short delay (TR 326 ~ 500) is then added before the next inversion.

On the sodium side, a non-selective pulse with a constant FA = 30° is applied at every other TR to reduce SAR and obtain a larger effective gradient spoiling moment for sodium (effective TR = 15 ms). The sodium acquisition continues throughout the whole scan without interruption (see Fig. 1).

The 3D simultaneous ¹H MRF/²³Na MRI sequence parameters were: FOV 240×240×168 mm³, ¹H 160×160×56 / ²³Na 84×84×56 matrix, ¹H 1.5×1.5×3 mm³ / ²³Na 2.85×2.85×3 mm³ resolution, ¹H TR = 7.5 ms / ²³Na TR = 15 ms, 1 slab, 6 shots per slab, total scan time 21 min.

All experiments were performed at 7T (MAGNETOM, Siemens, Erlangen, Germany) in accordance with the relevant institutional and national guidelines for informed consent from the subjects, using a 16-channel-Tx/Rx dual-tuned coil⁵ developed in-house.

The images were reconstructed offline in MATLAB (Mathworks, Natick, MA, USA). All images from each volunteer were co-registered and segmented in grey matter (GM), white matter (WM), and cerebral spinal fluid (CSF) with SPM 12 (UCL, London, UK)⁶. Then, we calculated the mean value and standard deviation (SD) of the PD, T₁, T₂, and ²³Na density for each tissue and scan. The PD was normalized by the value of the CSF and the ²³Na density by the mean eyes intensity.

For the statistical analysis, we estimated the intra-subject variance (the variance between results from scans of the same subject) and the inter-subject variance (the variance between results from different subjects) of each measure (mean and SD). Then, we computed the coefficient of variation⁷ (CV) as the square root of the intra-subject variance (intra-var) expressed as percentage of the mean, and the intra-class correlation⁸ (ICC) as the inter-subject variance (inter-var) divided by the sum of the intra and inter variances. The CV is considered an indicator of the utility of a measure for detecting within-subject changes over time. An ideal set of measurements has a CV equal to 0%. The ICC is a measure of the repeatability of the method, it has values between 0 and 1, where higher values are associated with more repeatable measurements.

Results & Discussion

Fig. 2 shows the ¹H maps and ²³Na images from the 3 scans of subject 1 and Fig. 3 shows the brain segmentation for subject 1. Finally, Table 1 summarizes the results of the statistical analysis.

The CV was lower than 10% in 23 of the 24 statistical results (mean and SD of all 4 measurements in 3 brain regions). The ICC was higher than 0.8 for 11 of the 12 T₁ and T₂ measurements, in the range of 0.65 and 0.77 for the PD measurements and in the range of 0.44 and 0.88 for the 6 ²³Na density. Hence, the method is highly repeatable for T₁ and T₂ measurements, while it is less robust for PD and ²³Na density. This can be due to T₁ and T₂ were calculated from a dictionary, making these measurements more robust under noisy conditions9, while PD and sodium density relied on local normalization for each scan (which make these maps more sensitive to B₀ and B₁ inhomogeneities).

Conclusion

In this work, we assessed the repeatability of PD, T₁, T₂, and ²³Na density measured from simultaneous 3D ¹H MRF/²³Na MRI acquisition over 21 scans of 7 different subjects in the brain at 7T. The overall repeatability was good, where the estimated CV were lower than 10% for 23 of the 24 statistical results and the ICC higher than 0.6 in 20 of the 24 statistical results.

Acknowledgements

The research reported in this publication was supported by the NIH/NIBIB grant R01 EB026456, and performed under the rubric of the Center for Advanced Imaging Innovation and Research, a NIBIB Biomedical Technology Resource Center (P41 EB017183).