An optimized MP2RAGE sequence for studying both brain and cervical spinal cord in a single acquisition at 3T.

Magnetization Prepared 2 Rapid Acquisition Gradient Echo (MP2RAGE) is a T1 mapping technique that has been used broadly on brain and recently on cervical spinal cord (cSC). The growing interest for combined investigation of brain and SC in numerous pathologies of the central nervous system such as multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS) and traumatic injuries, now brings about the need for optimization with regards to this specific investigation. This implies large spatial coverage with high spatial resolution and short acquisition time, high CNR and low B1+ sensitivity, as well as high reproducibility and robust post-processing tools for T1 quantification in different regions of brain and SC. In this work, a dedicated protocol (referred to as Pr-BSC) has been optimized for simultaneous brain and cSC T1 MP2RAGE acquisition at 3T. After computer simulation optimization, the protocol was applied for in vivo validation experiments and compared to previously published state of the art protocols focusing on either the brain (Pr-B) or the cSC (Pr-SC). Reproducibility and in-ROI standard deviations were assessed on healthy volunteers in the perspective of future clinical use. The mean T1 values, obtained by the Pr-BSC, in brain white, gray and deep gray matters were: (mean ± in-ROI SD) 792 ± 27 ms, 1339 ± 139 ms and 1136 ± 88 ms, respectively. In cSC, T1 values for white matter corticospinal, posterior sensory, lateral sensory and rubro/reticulospinal tracts were 902 ± 41 ms, 920 ± 35 ms, 903 ± 46 ms, 891 ± 41 ms, respectively, and 954 ± 32 ms for anterior and intermediate gray matter. The Pr-BSC protocol showed excellent agreement with previously proposed Pr-B on brain and Pr-SC on cSC, with very high inter-scan reproducibility (coefficients of variation of 0.52 ± 0.36% and 1.12 ± 0.62% on brain and cSC, respectively). This optimized protocol covering both brain and cSC with a sub-millimetric isotropic spatial resolution in one acquisition of less than 8 min, opens up great perspectives for clinical applications focusing on degenerative tissue such as encountered in MS and ALS.

Anterior fissure, central canal, posterior septum and more: New insights into the cervical spinal cord gray and white matter regional organization using T1 mapping at 7T.

T1 mapping lacks specificity toward a single particular biological feature, however it has the potential to discriminate spinal cord regional tissue organization and characterize tissue microstructural impairments occurring in neurodegenerative pathologies. In this exploratory work, T1 mapping of the cervical spinal cord with a 300-µm in-plane resolution was performed on fourteen healthy subjects at 7T, using the MP2RAGE sequence. Individual images from C1 to C7 vertebral levels provided a clear delineation of spinal cord anatomical details and substructures including motor columns within gray matter (GM) horns, anterior median fissure, central canal, ventral, lateral and dorsal white matter (WM) fasciculi, and posterior median septum. Group studies highlighted regional T1 differences between regions of interest so far hardly visible at lower spatial resolution. Two-dimensional averaged T1 maps and manual parcellation of GM and WM substructures were built based on these data. Benefiting from the very high spatial resolution achievable at ultra-high field for T1 mapping, this work contributes to improve the in vivo characterization of the cervical spinal cord. By allowing investigation within a wider range of functional regions, it also opens new perspectives for pathology diagnosis such as motor neuron disease, neuropathic pain or refined investigation of neurodegeneration.

Regional T1 mapping of the whole cervical spinal cord using an optimized MP2RAGE sequence.

The recently-proposed MP2RAGE sequence was purposely optimized for cervical spinal cord imaging at 3T. Sequence parameters were chosen to optimize gray/white matter T1 contrast with sub-millimetric resolution and scan-time < 10 min while preserving reliable T1 determination with minimal B1+ variation effects within a range of values compatible with pathologies and surrounding structures. Results showed good agreements with IR-based measurements, high MP2RAGE-based T1 reproducibility and preliminary evidences of age- and tract-related T1 variations in the healthy spinal cord.

While T1 measurements present multiple challenges (robustness, acquisition time), the recently proposed MP2RAGE sequence (magnetization-prepared two rapid acquisition gradient echoes) has opened new perspectives to characterize tissue microstructure changes occurring in a pathological or developmental context. Extensively used for brain studies, it was herein adapted to investigate the cervical spinal cord (SC) at 3 T. By integrating Bloch equations, the MP2RAGE sequence parameters were chosen to optimize SC gray matter/white matter (GM/WM) T1 contrast with sub-millimetric resolution, a scan time less than 10 min and a reliable T1 determination with minimal B1+ variation effect, within a range of values compatible with different pathologies and surrounding structures. The residual B1+ effect on T1 values was corrected using a look-up-table approach and B1+ mapping. The accuracy of B1+ -corrected T1 measurements was assessed on a phantom with respect to conventional inversion recovery. In vivo MP2RAGE acquisitions were performed on five young (28.8 ± 4.3 years old) and five elderly (60.2 ± 2.9 years old) volunteers and analyzed using a template-based approach. Phantom experiments led to high agreements between inversion-recovery spin-echo and MP2RAGE-based T1 values (R2  = 0.997). In vivo T1 values for cervical WM, anterior GM (aGM), posterior sensory tracts (PSTs) and lateral motor tracts (LMTs) were 917 ± 29 s, 934 ± 33 ms, 920 ± 37 ms and 877 ± 35 ms, respectively, with all subjects and cervical levels considered. Significant differences were observed between aGM and LMTs, and between LMTs and PSTs, in agreement with the literature. Repeated T1 measurements demonstrated high reproducibility of the MP2RAGE in the SC (variation coefficient < 5% in all regions of interest). Finally, preliminary assessment of age-related SC tissue microstructure variation additionally showed evidence of SC atrophy and slight trends of T1 decrease with age in all regions. Overall, this study shows that fast, robust and accurate sub-millimetric resolution T1 mapping in the cervical SC using the MP2RAGE sequence is possible, paving the way for future multi-centric and longitudinal clinical studies investigating the pathological cord.

Spatial distribution of multiple sclerosis lesions in the cervical spinal cord.

Spinal cord lesions detected on MRI hold important diagnostic and prognostic value for multiple sclerosis. Previous attempts to correlate lesion burden with clinical status have had limited success, however, suggesting that lesion location may be a contributor. Our aim was to explore the spatial distribution of multiple sclerosis lesions in the cervical spinal cord, with respect to clinical status. We included 642 suspected or confirmed multiple sclerosis patients (31 clinically isolated syndrome, and 416 relapsing-remitting, 84 secondary progressive, and 73 primary progressive multiple sclerosis) from 13 clinical sites. Cervical spine lesions were manually delineated on T2- and T2*-weighted axial and sagittal MRI scans acquired at 3 or 7 T. With an automatic publicly-available analysis pipeline we produced voxelwise lesion frequency maps to identify predilection sites in various patient groups characterized by clinical subtype, Expanded Disability Status Scale score and disease duration. We also measured absolute and normalized lesion volumes in several regions of interest using an atlas-based approach, and evaluated differences within and between groups. The lateral funiculi were more frequently affected by lesions in progressive subtypes than in relapsing in voxelwise analysis (P < 0.001), which was further confirmed by absolute and normalized lesion volumes (P < 0.01). The central cord area was more often affected by lesions in primary progressive than relapse-remitting patients (P < 0.001). Between white and grey matter, the absolute lesion volume in the white matter was greater than in the grey matter in all phenotypes (P < 0.001); however when normalizing by each region, normalized lesion volumes were comparable between white and grey matter in primary progressive patients. Lesions appearing in the lateral funiculi and central cord area were significantly correlated with Expanded Disability Status Scale score (P < 0.001). High lesion frequencies were observed in patients with a more aggressive disease course, rather than long disease duration. Lesions located in the lateral funiculi and central cord area of the cervical spine may influence clinical status in multiple sclerosis. This work shows the added value of cervical spine lesions, and provides an avenue for evaluating the distribution of spinal cord lesions in various patient groups.

Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks.

The spinal cord is frequently affected by atrophy and/or lesions in multiple sclerosis (MS) patients. Segmentation of the spinal cord and lesions from MRI data provides measures of damage, which are key criteria for the diagnosis, prognosis, and longitudinal monitoring in MS. Automating this operation eliminates inter-rater variability and increases the efficiency of large-throughput analysis pipelines. Robust and reliable segmentation across multi-site spinal cord data is challenging because of the large variability related to acquisition parameters and image artifacts. In particular, a precise delineation of lesions is hindered by a broad heterogeneity of lesion contrast, size, location, and shape. The goal of this study was to develop a fully-automatic framework - robust to variability in both image parameters and clinical condition - for segmentation of the spinal cord and intramedullary MS lesions from conventional MRI data of MS and non-MS cases. Scans of 1042 subjects (459 healthy controls, 471 MS patients, and 112 with other spinal pathologies) were included in this multi-site study (n = 30). Data spanned three contrasts (T1-, T2-, and T2∗-weighted) for a total of 1943 vol and featured large heterogeneity in terms of resolution, orientation, coverage, and clinical conditions. The proposed cord and lesion automatic segmentation approach is based on a sequence of two Convolutional Neural Networks (CNNs). To deal with the very small proportion of spinal cord and/or lesion voxels compared to the rest of the volume, a first CNN with 2D dilated convolutions detects the spinal cord centerline, followed by a second CNN with 3D convolutions that segments the spinal cord and/or lesions. CNNs were trained independently with the Dice loss. When compared against manual segmentation, our CNN-based approach showed a median Dice of 95% vs. 88% for PropSeg (p ≤ 0.05), a state-of-the-art spinal cord segmentation method. Regarding lesion segmentation on MS data, our framework provided a Dice of 60%, a relative volume difference of -15%, and a lesion-wise detection sensitivity and precision of 83% and 77%, respectively. In this study, we introduce a robust method to segment the spinal cord and intramedullary MS lesions on a variety of MRI contrasts. The proposed framework is open-source and readily available in the Spinal Cord Toolbox.

Region-specific impairment of the cervical spinal cord (SC) in amyotrophic lateral sclerosis: A preliminary study using SC templates and quantitative MRI (diffusion tensor imaging/inhomogeneous magnetization transfer).

In this preliminary study, our objective was to investigate the potential of high-resolution anatomical imaging, diffusion tensor imaging (DTI) and conventional/inhomogeneous magnetization transfer imaging [magnetization transfer (MT)/inhomogeneous magnetization transfer (ihMT)] at 3 T, analyzed with template-extracted regions of interest, to measure the atrophy and structural changes of white (WM) and gray (GM) matter spinal cord (SC) occurring in patients with amyotrophic lateral sclerosis (ALS). Ten patients with ALS and 20 age-matched healthy controls were recruited. SC GM and WM areas were automatically segmented using dedicated templates. Atrophy indices were evaluated from T2 *-weighted images at each vertebral level from cervical C1 to C6. DTI and ihMT metrics were quantified within the corticospinal tract (CST), posterior sensory tract (PST) and anterior GM (aGM) horns at the C2 and C5 levels. Clinical disabilities of patients with ALS were evaluated using the Revised ALS Functional Rating Scale, upper motor neuron (UMN) and Medical Research Council scorings, and correlated with MR metrics. Compared with healthy controls, GM and WM atrophy was observed in patients with ALS, especially at lower cervical levels, where a strong correlation was also observed between GM atrophy and the UMN score (R = -0.75, p = 0.05 at C6). Interestingly, a significant decrease in ihMT ratio was found in all regions of interest (p < 0.0008), fractional anisotropy (FA) and MT ratios decreased significantly in CST, especially at C5 (p < 0.005), and λ// (axial diffusivity) decreased significantly in CST (p = 0.0004) and PST (p = 0.003) at C2. Strong correlations between MRI metrics and clinical scores were also found (0.47 < |R| < 0.87, p < 0.05). Altogether, these preliminary results suggest that high-resolution anatomical imaging and ihMT imaging, in addition to DTI, are valuable for the characterization of SC tissue impairment in ALS. In this study, in addition to an important SC WM demyelination, we also observed, for the first time in ALS, impairments of cervical aGM.