Ultrahigh-b diffusion-weighted imaging for quantitative evaluation of myelination in shiverer mouse spinal cord.

PURPOSE: To perform a quantitative evaluation of myelination on WT and myelin-deficient (shiverer) mouse spinal cords using ultrahigh-b diffusion-weighted imaging (UHb-DWI). METHODS: UHb-DWI of ex vivo on spinal cord specimens of two shiverer (C3HeB/FeJ-shiverer, homozygous genotype for MbPshi ) and six WT (Black Six, C3HeB/FeJ) mice were acquired using 3D multishot diffusion-weighted stimulated-echo EPI, a homemade RF coil, and a small-bore 7T MRI system. Imaging was performed in transaxial plane with 75 × 75 µm2 in-plane resolution, 1-mm-slice thickness, and radial DWI using bmax = 42,890 s/mm2 . Histological evaluation was performed on upper thoracic sections using optical and transmission electron microscopy. Numerical Monte Carlo simulations (MCSs) of water diffusion were performed to facilitate interpretation of UHb-DWI signal-b curves. RESULTS: The white matter ultrahigh-b radial DWI (UHb-rDWI) signal-b curves of WT mouse cords behaved biexponentially with high-b diffusion coefficient DH < 0.020 × 10-3 mm2 /s. However, as expected with less myelination, the signal-b of shiverer mouse cords behaved monoexponentially with significantly greater DH = 0.162 × 10-3 , 0.142 × 10-3 , and 0.164 × 10-3 mm2 /s at anterodorsal, posterodorsal, and lateral columns, respectively. The axial DWI signals of all mouse cords behaved monoexponentially with D = (0.718-1.124) × 10-3 mm2 /s. MCS suggests that these elevated DH are mainly induced by increased water exchange at the myelin sheath. Microscopic results were consistent with the UHb-rDWI findings. CONCLUSION: UHb-DWI provides quantitative differences in myelination of spinal cords from myelin-deficit shiverer and WT mice. UHb-DWI may become a powerful tool to evaluate myelination in demyelinating disease models that may translate to human diseases, including multiple sclerosis.

High-b diffusivity of MS lesions in cervical spinal cord using ultrahigh-b DWI (UHb-DWI).

PURPOSE: The purpose of this study was to investigate UHb-rDWI signal in white matter tracts of the cervical spinal cord (CSC) and compare quantitative values between healthy control WM with both MS NAWM and MS WM lesions. METHODS: UHb-rDWI experiments were performed on (a) 7 MS patients with recently active or chronic lesions in CSC and on (b) 7 healthy control of similar age range and gender distribution to MS subjects. All MRI data were acquired using clinical 3T MRI system. Axial high-b diffusion images were acquired using 2D single-shot DW stimulated EPI with reduced FOV and a CSC-dedicated 8 channel array coil. High-b diffusion coefficient DH was estimated by fitting the signal-b curve to a double or single-exponential function. RESULTS: The high-b diffusivity DH values were measured as (0.767 ± 0.297) × 10-3 mm2/s in the posterior column lesions, averaged over 6 MS patients, and 0.587 × 10-3 mm2/s in the corticospinal tract for another patient. The averaged DH values of the 7 healthy volunteers from the posterior and lateral column were (0.0312 ± 0.0306) × 10-3 and (0.0505 ± 0.0205) × 10-3 mm2/s, respectively. UHb-rDWI signal-b curves of the MS patients revealed to noticeably behave differently to that of the healthy controls. The patient signal-b curves decayed with greater high-b decay constants to reach lower signal intensities relative to signal-b curves of the healthy controls. CONCLUSION: UHb-DWI of the CSC reveals a marked difference in signal-b-curves and DH values in MS lesions compared to NAWM and healthy control WM. Based on physical principles, we interpret these altered observations of quantitative diffusion values to be indicative of demyelination. Further studies in animal models will be required to fully interpret UHb-DWI quantitative diffusion values during demyelination and remyelination.

Diffusion MRI using two-dimensional single-shot radial imaging (2D ss-rDWI) with variable flip angle and random view ordering.

PURPOSE: The main objective of this study is to develop a 2D single-shot radial-DWI (2D ss-rDWI) technique to reduce motion artifacts and geometric distortion in DW images. METHOD: A diffusion-preparation module is developed and applied prior to the data acquisition. Because the diffusion-prepared longitudinal magnetization is measured over multiple RF excitations in each shot, 2D ss-rDWI is subject to low signal-to-noise ratio (SNR). We used variable-flip angle (VFA), random view ordering (RVO), and sliding spokes, and compared the performances to constant flip angle (CFA), smooth view ordering (SVO), and identical spoke averaging, respectively. For each technique, we performed numerical simulation and MRI experiments on a fluid phantom as well as in-vivo human brain studies with a 3 T MRI system. RESULTS: Using VFA, optimal SNR was acquired for 2D ss-rDWI. Using SVO, the high signal is clustered at specific quadrant in 2D k-space: the first quadrant using high initial flip angle or the last quadrant using the low flip angle. This clustered signal in k-space led to geometric distortion in image space. 2D ss-rDWI using RVO spreads the high signaled spokes over all angular directions and removes the view-order-related distortion. The in-vivo images using 2D ss-rDWI with VFA and RVO show no geometric distortion at the skull base brain, but greatly reduced SNR compared with those using 2D ss-DWEPI. CONCLUSION: 2D ss-rDWI is optimized by using VFA with RVO. The resultant DWI using 2D ss-rDWI is insensitive to motion-induced artifacts and geometric distortion. Even with low SNR, it may be useful for DWI of organs limited by severe susceptibility-induced geometric distortion.

Ultra-high-b radial diffusion-weighted imaging (UHb-rDWI) of human cervical spinal cord.

BACKGROUND: Injury in the cervical spinal cord (CSC) can lead to varying degrees of neurologic deficit and persistent disability. Diffusion tensor imaging (DTI) is a promising method to evaluate white matter integrity and pathology. However, the conventional DTI results are limited with respect to the specific details of neuropathology and microstructural architecture. In this study we used ultrahigh-b radial-DWI (UHb-rDWI) with b-values ranging from 0 to ∼7500 s/mm2 and calculated decay constant (DH ) at the high b-values, which gives much deeper insight about the microscopic environment of CSC white matter. PURPOSE: To evaluate a novel diffusion MRI, UHb-rDWI technique for imaging of the CSC. STUDY TYPE: Longitudinal. SUBJECTS: Four healthy controls, each scanned twice. FIELD STRENGTH/SEQUENCE: 3T/2D single shot diffusion-weighted stimulated echo planar imaging with reduced field of view. ASSESSMENT: The signal from each pixel of b0 (b = 0) and b-value (b ≠ 0) images were fitted to a biexponential function and normalized. The signal-b curve is obtained by dividing the latter curve by the former. DH was obtained from the curve at b >4000 s/mm2 . A Monte-Carlo Simulation (MCS) was performed to investigate how DH changes upon the increased water-exchange at the CSC. RESULTS: The signal-b curves plotted at multiple levels of healthy CSC are almost identical on two successive scans and show a biexponential decay behavior: fast exponential decay at lower b-values and much slower decay at UHb-values. The mean values of DH were measured as (0.0607 ± 0.02531) ×10-3 and (0.0357 ± 0.02072) ×10-3 s/mm2 at the lateral funiculus and posterior column, respectively. MCS of diffusion MRI shows that the DH is elevated by increased water exchange between the intra- and extraaxonal spaces. DATA CONCLUSION: UHb-rDWI signal-b plots of the normal CSC were highly reproducible on successive scans and their biexponential decay behavior can be used to characterize normal spinal white matter. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;49:204-211.

Synchronous Radial 1H and 23Na Dual-Nuclear MRI on a Clinical MRI System, Equipped With a Broadband Transmit Channel.

The purpose of this work was to synchronously acquire proton (1H) and sodium (23Na) image data on a 3T clinical MRI system within the same sequence, without internal modification of the clinical hardware, and to demonstrate synchronous acquisition with 1H/23Na-GRE imaging with Cartesian and radial k-space sampling. Synchronous dual-nuclear imaging was implemented by: mixing down the 1H signal so that both the 23Na and 1H signal were acquired at 23Na frequency by the conventional MRI system; interleaving 1H/23Na transmit pulses in both Cartesian and radial sequences; and using phase stabilization on the 1H signal to remove mixing effects. The synchronous 1H/23Na setup obtained images in half the time necessary to sequentially acquire the same 1H and 23Na images with the given setup and parameters. Dual-nuclear hardware and sequence modifications were used to acquire 23Na images within the same sequence as 1H images, without increases to the 1H acquisition time. This work demonstrates a viable technique to acquire 23Na image data without increasing 1H acquisition time using minor additional custom hardware, without requiring modification of a commercial scanner with multinuclear capability.