Elsevier

NeuroImage

Volume 67, 15 February 2013, Pages 64-76
NeuroImage

Improved in vivo diffusion tensor imaging of human cervical spinal cord

https://doi.org/10.1016/j.neuroimage.2012.11.014Get rights and content

Abstract

We describe a cardiac gated high in-plane resolution axial human cervical spinal cord diffusion tensor imaging (DTI) protocol. Multiple steps were taken to optimize both image acquisition and image processing. The former includes slice-by-slice cardiac triggering and individually tiltable slices. The latter includes (i) iterative 2D retrospective motion correction, (ii) image intensity outlier detection to minimize the influence of physiological noise, (iii) a non-linear DTI estimation procedure incorporating non-negative eigenvalue priors, and (iv) tract-specific region-of-interest (ROI) identification based on an objective geometry reference. Using these strategies in combination, radial diffusivity (λ) was reproducibly measured in white matter (WM) tracts (adjusted mean [95% confidence interval] = 0.25 [0.22, 0.29] μm2/ms), lower than previously reported λ values in the in vivo human spinal cord DTI literature. Radial diffusivity and fractional anisotropy (FA) measured in WM varied from rostral to caudal as did mean translational motion, likely reflecting respiratory motion effect. Given the considerable sensitivity of DTI measurements to motion artifact, we believe outlier detection is indispensable in spinal cord diffusion imaging. We also recommend using a mixed-effects model to account for systematic measurement bias depending on cord segment.

Graphical abstract

Highlights

► A comprehensive tract-specific DTI protocol for human cervical spinal cord. ► Small radial diffusivity in WM highlighted the effect of optimization. ► Respiratory motion likely biased DTI parameters in spinal cord. ► Mixed-effects model accounted for systematic measurement bias. ► Outlier removal is indispensable for reproducible spinal cord diffusion imaging.

Introduction

Spinal cord damage is a major cause of clinical disability in multiple sclerosis (MS), neuromyelitis optica (NMO), amyotrophic lateral sclerosis (ALS) and trauma. The spinal cord contains white matter (WM) tracts that carry information to and from the brain. The lateral corticospinal tracts (CST), located in the lateral portion of the cord, convey motor commands from the brain to all spinal segments; the posterior columns (PC), located in the dorsal portion of the cord, relay sensory impulses upwards to nuclei in the brainstem. The ability to evaluate tissue integrity within these specific WM tracts would have immediate clinical impact.

Conventional spinal cord magnetic resonance imaging (MRI), using T2-weighted (T2W), T1-weighted (T1W), and short-TI inversion recovery (STIR) sequences, can reveal the location of lesions that are either hyper- or hypo-intense. However, such qualitative assessments are not specific with respect to the underlying disease process leading to white matter pathology. Diffusion tensor imaging (DTI, Basser and Pierpaoli, 1996) improves pathologic specificity through the quantitative directional diffusivities, which measure water diffusion parallel (i.e., axial or longitudinal diffusivity equals λ1 or λ||) and perpendicular (i.e., radial or perpendicular diffusivity equals λ = λ2 + λ3/2) to the WM tracts (Song et al., 2002, Song et al., 2003). These parameters have been shown to reflect axon and myelin damage in mouse models of spinal cord injury and disease (Budde et al., 2009, Kim et al., 2006).

However, quantitative in vivo DTI of the human spinal cord (Clark and Werring, 2002, Ellingson et al., 2008, Maier, 2007, Maier and Mamata, 2005) is challenging for several reasons (Barker, 2001): (i) The cord has a small cross-sectional area; (ii) magnetic field inhomogeneities from nearby vertebrae cause image distortions; and (iii) cerebrospinal fluid (CSF) pulsations, cardiac pulsation in surrounding vessels, and respiratory motion generate significant motion artifacts in the anterior-posterior and rostro-caudal directions. Thus, limited signal-to-noise ratio (SNR) resulting from physiological artifacts as well as thermal noise is a major concern in high in-plane resolution axial spinal cord DTI acquisition.

Reduced field-of-view (rFOV) sequences (Dowell et al., 2009, Finsterbusch, 2009, Jeong et al., 2005, Jeong et al., 2006, Kim et al., 2010, Saritas et al., 2008, Wheeler-Kingshott et al., 2002a, Wilm et al., 2007, Wilm et al., 2009, Xu et al., 2010) have shown promise in producing high quality, in vivo human spinal cord DTI. However, despite improvements in image quality and spatial resolution, tract-specific diffusion quantification within the individual spinal cord WM tracts has not been entirely satisfactory. Few studies have employed cardiac gating (Cohen-Adad et al., 2011a, Kharbanda et al., 2006, Summers et al., 2006, Wheeler-Kingshott et al., 2002a) to minimize the effect of CSF pulsations or have implemented outlier detection routines (Freund et al., 2010) as quality control measures. Consequently, overestimation of water diffusivities is likely prevalent in the human spinal cord DTI literature. In addition, motion correction, which depends on image registration, is challenging in rFOV spinal cord DTI because of limited tissue contrast and the small FOV.

Here we propose a slice-by-slice cardiac gated, rFOV cervical spinal cord DTI protocol. Features of this protocol include (i) iterative (Shimony et al., 2006) 2D image registration (Smith et al., 2010) optimized for rFOV spinal cord diffusion weighted (DW) images; (ii) outlier detection (Chang et al., 2005) to minimize the influence of physiological noise; (iii) a diffusion tensor estimation procedure incorporating positive eigenvalue priors, similar to (Cohen-Adad et al., 2008, Ducreux et al., 2007, Onu et al., 2010); and (iv) tract-specific region of interest (ROI) demarcation based on known anatomy (Klawiter et al., 2012, Xu et al., 2010). We offer this protocol as a comprehensive solution to the aforementioned challenges of tract-specific quantitative DTI of in vivo human spinal cord.

Section snippets

Subjects

Eighteen neurologically normal volunteers (median age 34 ± 13 years, range 19–70 years; 8 females and 10 males) participated in this study. Each participant provided a written informed consent. All aspects of the study were approved by the Washington University Human Research Protection Office. Eight of the volunteers (1 subject, 4 times; 1 subject, 3 times, and 6 subjects, 2 times) were re-scanned over a 2-year period. Subjects were instructed to breathe and swallow normally during the entire data

Results

The average number of outlier images per slice was 6 ± 5% (mean ± sd). The number of outlier images varied substantially between slices and subjects with a trend of more outlier images towards the caudal slices. The outliers identified as excessive translation (i.e., failed registration) before the initial DTI fitting accounted for less than 1% of the total number of outlier images. In DTI data processed without outlier rejection there were appreciable overestimations of λ (Fig. 5A), λ|| (Fig. 5

Discussion

Starting with the improved image quality and spatial resolution offered by rFOV diffusion imaging (Dowell et al., 2009, Finsterbusch, 2009, Jeong et al., 2005, Jeong et al., 2006, Kim et al., 2010, Saritas et al., 2008, Wheeler-Kingshott et al., 2002a, Wilm et al., 2007, Wilm et al., 2009), we employed slice-by-slice cardiac triggering and individually tiltable slices in spinal cord DTI acquisition (Klawiter et al., 2012, Xu et al., 2010). We further implemented several post-processing steps to

Acknowledgments

We thank Mark Jenkinson (fMRI group, Oxford, UK) for providing a 2D translation-only (DOF = 2) schedule file for FSL FLIRT registration. We are also grateful to Larry Bretthorst (Biomedical Magnetic Resonance Laboratory, Washington University School of Medicine) for instrumental discussions about the implementation of Bayesian positive priors in the non-linear DTI fitting method. We appreciate the technical and scanning assistance of Glenn J. Foster, Mark A. Nolte and Scott M. Love (Center for

References (93)

  • M. Jenkinson et al.

    Improved optimization for the robust and accurate linear registration and motion correction of brain images

    Neuroimage

    (2002)
  • J. Kim et al.

    Detecting axon damage in spinal cord from a mouse model of multiple sclerosis

    Neurobiol. Dis.

    (2006)
  • J.H. Kim et al.

    Diffusion tensor imaging of mouse brain stem and cervical spinal cord

    J. Neurosci. Methods

    (2009)
  • E.C. Klawiter et al.

    Radial diffusivity predicts demyelination in ex vivo multiple sclerosis spinal cords

    Neuroimage

    (2011)
  • H. Lundell et al.

    Distribution of collateral fibers in the monkey cervical spinal cord detected with diffusion-weighted magnetic resonance imaging

    Neuroimage

    (2011)
  • S. Maier

    Examination of spinal cord tissue architecture with magnetic resonance diffusion tensor imaging

    Neurotherapeutics

    (2007)
  • H. Mamata et al.

    Collateral nerve fibers in human spinal cord: visualization with magnetic resonance diffusion tensor imaging

    Neuroimage

    (2006)
  • J.D. Power et al.

    Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion

    Neuroimage

    (2012)
  • K. Shmueli et al.

    Low-frequency fluctuations in the cardiac rate as a source of variance in the resting-state fMRI BOLD signal

    Neuroimage

    (2007)
  • S.-K. Song et al.

    Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water

    Neuroimage

    (2002)
  • S.-K. Song et al.

    Diffusion tensor imaging detects and differentiates axon and myelin degeneration in mouse optic nerve after retinal ischemia

    Neuroimage

    (2003)
  • A. Volk et al.

    Chemical shift-specific slice selection. A new method for chemical shift imaging at high magnetic field

    J. Magn. Reson.

    (1987)
  • H.U. Voss et al.

    Fiber tracking in the cervical spine and inferior brain regions with reversed gradient diffusion tensor imaging

    Magn. Reson. Imaging

    (2006)
  • L. Walker et al.

    Effects of physiological noise in population analysis of diffusion tensor MRI data

    Neuroimage

    (2011)
  • C.A.M. Wheeler-Kingshott et al.

    Investigating cervical spinal cord structure using axial diffusion tensor imaging

    Neuroimage

    (2002)
  • S. Akselrod et al.

    Power spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control

    Science

    (1981)
  • N. Alperin et al.

    Hemodynamically independent analysis of cerebrospinal fluid and brain motion observed with dynamic phase contrast MRI

    Magn. Reson. Med.

    (1996)
  • V. Arsigny et al.

    Log-Euclidean metrics for fast and simple calculus on diffusion tensors

    Magn. Reson. Med.

    (2006)
  • Y. Assaf et al.

    Axcaliber: a method for measuring axon diameter distribution from diffusion MRI

    Magn. Reson. Med.

    (2008)
  • D. Bates et al.

    lme4: linear mixed-effects models using S4 classes

    (2011)
  • J.M. Bland et al.

    Measuring agreement in method comparison studies

    Stat. Methods Med. Res.

    (1999)
  • M.D. Budde et al.

    Toward accurate diagnosis of white matter pathology using diffusion tensor imaging

    Magn. Reson. Med.

    (2007)
  • M.D. Budde et al.

    Axial diffusivity is the primary correlate of axonal injury in the experimental autoimmune encephalomyelitis spinal cord: a quantitative pixelwise analysis

    J. Neurosci.

    (2009)
  • L. Chang et al.

    RESTORE: robust estimation of tensors by outlier rejection

    Magn. Reson. Med.

    (2005)
  • L.-C. Chang et al.

    Informed RESTORE: a method for robust estimation of diffusion tensor from low redundancy datasets in the presence of physiological noise artifacts

    Magn. Reson. Med.

    (2012)
  • O. Ciccarelli et al.

    Spinal cord spectroscopy and diffusion-based tractography to assess acute disability in multiple sclerosis

    Brain

    (2007)
  • C.A. Clark et al.

    Diffusion tensor imaging in spinal cord: methods and applications—a review

    NMR Biomed.

    (2002)
  • Y. Cohen et al.

    High b-value q-space analyzed diffusion-weighted MRS and MRI in neuronal tissues – a technical review

    NMR Biomed.

    (2002)
  • J. Cohen-Adad et al.

    Distortion correction in spinal cord DTI: what's the best approach?

  • J. Cohen-Adad et al.

    32‐Channel RF coil optimized for brain and cervical spinal cord at 3 T

    Magn. Reson. Med.

    (2011)
  • G.C. DeLuca et al.

    Axonal loss in multiple sclerosis: a pathological survey of the corticospinal and sensory tracts

    Brain

    (2004)
  • R Development Core Team

    R: a language and environment for statistical computing

    (2011)
  • N.G. Dowell et al.

    Contiguous-slice zonally oblique multislice (CO-ZOOM) diffusion tensor imaging: examples of in vivo spinal cord and optic nerve applications

    J. Magn. Reson. Imaging

    (2009)
  • B.M. Ellingson et al.

    Morphology and Morphometry of Human Chronic Spinal Cord Injury Using Diffusion Tensor Imaging and Fuzzy Logic

    Ann. Biomed. Eng.

    (2007)
  • B.M. Ellingson et al.

    Diffusion tensor magnetic resonance imaging in spinal cord injury

    Concepts Magn. Reson. A

    (2008)
  • J.A.D. Farrell et al.

    Effects of signal‐to‐noise ratio on the accuracy and reproducibility of diffusion tensor imaging–derived fractional anisotropy, mean diffusivity, and principal eigenvector measurements at 1.5T

    J. Magn. Reson. Imaging

    (2007)
  • Cited by (0)

    1

    Dr. Xu is now affiliated with Mount Sinai School of Medicine.

    View full text