4.2. The Lumbar Cord
4.2.2. Materials and Methods
The local Institutional Review Board approved this study, and signed informed consent was obtained prior to the examination. Data were obtained from four healthy volunteers (1 male, mean age 26.5±5.5 years) on a 3T Philips Achieva scanner (Philips Healthcare, Best, The Netherlands). A quadrature body coil was used for excitation and a 6 channel CTL whole spine clinical standard phased array was used for signal reception. The field-of-view (FOV) was centered on the thoracolumbar bulge between the T12-L1 vertebral bodies (see Figure 4.5), and spanned, at minimum, the T11 to L2 vertebral levels in all subjects.
Figure 4.5. (a.) Low-resolution, multiple RF power raw MT-weighted data for the middle slice of the FOV (b.) in a single volunteer. (b.) A T2-weighted sagittal slice over the full spinal cord. The blue box indicates the scan volume for all quantitative data. (c.) Anatomical mFFE and high-resolution MT-weighted data at the same slice as that in (a.). Notice improved GM/WM contrast. MT-weighted images with ROIs for the GM (red) and WM (green) are shown as well. (d.) R1, and associated full fit PSR from the data in (a.).
Two MT protocols were performed: 1) a low spatial resolution acquisition (1.5 x 1.5mm2) at 8 offsets (∆) and 2 powers (RF) with a “full-fit” analysis (78,82) and 2) a high-resolution acquisition (0.65 x 0.65 mm2) at 1 offset and power with a “single-point” analysis (37). For the qMT experiments, MT-weighted images were acquired using a 3D MT-prepared spoiled gradient echo sequence (36), and TR/TE/ = 55 ms/4.5 ms/15˚, with oversampling in the P/A direction of 30/60 mm. Nominal resolution for the low resolution, full model acquisition was 1.5 x 1.5 mm2 in-plane (reconstructed to 0.6 x 0.6mm2), and 0.65 x 0.65 mm2 for the high-resolution, single- point acquisition, over 8 slices (5 mm reconstructed slice thickness). Other parameters were:
FOV = 150 x 150 mm2, and 3 and 4 signal averages for the full-fit and single point acquisitions, respectively. MT weighting was achieved using a 20-ms, single-lobed sinc pulse with Gaussian apodization, RF = 450˚ and 900˚, and offset frequencies (∆ω) = 1, 1.5, 2, 2.5, 8, 16, 32, 100 kHz (chosen to approximately logarithmically sample the expected MT z-spectra (37)). MT weighting for the high-resolution, single-point data was achieved using the same pulse as the full-fit data, but at = 2.5 and 100 kHz, and an RF of 900˚. The total scan time for the low-resolution, full- fit acquisition was 19 minutes, 20 seconds, and the scan time for the single-point acquisition was 7 minutes, 23 seconds.
To correct for B0 inhomogeneities across the spinal cord, B0 maps were acquired using fast 3D techniques: dual-TE GRE with TR/TE1/TE2 = 50/4.6/6.9 ms and = 60˚. T1 mapping was performed using a multiple flip angle (MFA) acquisition with TR/TE = 20/4.6 ms and = 5, 10, 15, 20, 25, 30˚. A high-resolution multi-echo gradient echo (mFFE) scan was also performed and all echoes were averaged to generate a high contrast reference image for registration (163). The mFFE was obtained with TR/TE/∆TE = 700/7.2/9.3 ms and = 33˚, with a slice thickness of 3 mm, and a gap of 2 mm. Nominal in-plane resolution was 0.65 x 0.65 mm2. Acquisition times were 34 seconds for the B0 map, 2 minute 30 seconds for the T1 map, and 5 minutes 30 seconds for the mFFE, resulting in a total acquisition time of ≈36 minutes.
Image Processing
All data analyses were performed in MATLAB R2015b (MathWorks, Natick, MA). Prior to data fitting, all images were cropped to an area immediately around the spinal cord and co-registered to the mFFE volume using the FLIRT package from FSL v5.0.2.1 (FMRIB, Oxford, UK) (145,146).
The co-registration was limited to translation and rotation (±5˚) in-plane only (i.e. translation in x and y, and rotation about the z-axis). Following co-registration, qMT parameter maps were generated for each volunteer and patient using the full-fit qMT model described in Yarnykh (78)
& Yarnykh and Yuan (82). This model contains six independent parameters: R1m, R1f, T2m, T2f, PSR
= M0m/M0f, and kmf = kfm/PSR. It has been shown that the signal dependence on R1m is weak (19,34); therefore, it was set to R1f for fitting purposes. R1obs (1/T1obs) maps were reconstructed by regressing MFA data to the spoiled gradient echo signal equation in the steady-state (148).
The resulting maps were then used to determine the parameter R1f, as described by Yarnykh (78)
& Yarnykh and Yuan (82). The remaining parameters were estimated for each voxel by fitting the full-fit qMT data to the two-pool MT model (78,82). For all fitting, the nominal offset frequency was corrected in each voxel using B0.
It has been shown that T2m, kmf, and the product T2fR1f can be fixed during the fitting process because they exhibit relatively constant values across tissues(82). Note that T2f can be determined from the constrained T2fR1f value and an R1f estimate from the MFA data, while kfm
can be determined using the first-order mass action kinetics (127). Thus, these constraints result in a model where PSR is the only free parameter. To estimate reasonable fixed parameter values for the single-point qMT analysis, T2fR1f, kmf, and T2m were estimated over all voxels from all subjects, and histograms were created over the spinal cord for all slices from the full-fit analysis.
The median of each histogram was chosen to enter into the single-point qMT analysis. Then, using the median values of the constraints and the optimized offset and power (determined in Chapter 3), the high-resolution, single-point data were analyzed to generate high-resolution PSR maps.
Mean PSR values for the single-point and the full-fit scans were calculated from regions of interest (ROI) for each slice in the GM and WM, as shown in Figure 4.5c. ROIs were placed manually using MIPAV (NIH, Bethesda, MD) for each slice of each subject. The mean values from each ROI were then compared between the full fit and single point data, and between the lumbar cord and cervical cord data (see Chapter 3).