Elsevier

NeuroImage

Volume 67, 15 February 2013, Pages 193-202
NeuroImage

Imaging neural architecture of the brain based on its multipole magnetic response

https://doi.org/10.1016/j.neuroimage.2012.10.050Get rights and content

Abstract

Although magnetic fields interact weakly with biological tissues, at high fields, this interaction is sufficiently strong to cause measurable shifts in the Larmor frequency among various tissue types. While measuring frequency shift and its anisotropy has enabled NMR spectroscopy to determine structures of large molecules, MRI has not been able to fully utilize the vast information existing in the frequency to elucidate tissue microstructure. Using a multipole analysis of the complex MRI signal in the Fourier spectral space, we developed a fast and high-resolution method that enables the quantification of tissue's magnetic response with a set of magnetic susceptibility tensors of various ranks. The Fourier spectral space, termed p-space, can be generated by applying field gradients or equivalently by shifting the k-space data in various directions. Measuring these tensors allows the visualization and quantification of tissue architecture. We performed 3D whole-brain multipole susceptibility tensor imaging in simulation, on intact mouse brains ex vivo and on human brains in vivo. We showed that these multipole susceptibility tensors can be used to image orientations of ordered white matter fibers. These experiments demonstrate that multipole tensor analysis may enable practical mapping of tissue microstructure in vivo without rotating subject or magnetic field.

Graphical abstract

Highlights

► Magnetic response of the brain is described with multipole tensors ► Multipole tensor characterizes anisotropic magnetic response without any rotation ► Simulated axons are used to verify the properties of multipole tensors ► Fiber orientations of mouse brains are imaged with multipole tensors ► Fiber orientations of human brains are imaged in vivo

Introduction

Magnetic fields can penetrate deep into the body since they interact with biological molecules weakly as evidenced by the routine application of MRI in human bodies. Because of this weak interaction, MRI has traditionally relied on the amplitude of the nuclear magnetization from the very beginning to generate tissue contrast (Lauterbur, 1973). However, at high fields, interaction between magnetic field and the orbital electrons of biomolecules may introduce a measurable perturbation on the resonance frequency of surrounding water protons. This perturbation in turn reflects the molecular content and microstructure of the tissue. A notable example is the relative frequency shift between gray and white matter and between layers of the cortex which is thought to originate from variations of magnetic susceptibility (Duyn et al., 2007, Rauscher et al., 2005). Although frequency shift has provided a new image contrast for MRI, utilizing this contrast to infer neural architecture and brain structural connectivity remain challenging.

One potential way to fully utilize this frequency is to borrow techniques from NMR spectroscopy. Indeed, measuring frequency shift has been instrumental in NMR spectroscopy for probing molecular structure. While high-resolution NMR techniques provide a wealth of information (de Beer et al., 1994, Otting et al., 1990, Tolman et al., 1995, van Zijl et al., 1984), adapting those techniques to high-resolution imaging is not yet possible. The difficulty is partially due to low sensitivity, limited scan time and vastly more complex physiological conditions encountered in volumetric brain imaging. Because of these difficulties, frequency shift measured by MRI has been limited to the zero-th order information, i.e. the mean frequency of a whole voxel (Dixon, 1984, Glover and Schneider, 1991, Haacke et al., 1995, Rauscher et al., 2005, Weisskoff and Kiihne, 1992). Higher-order information such as susceptibility anisotropy of dipoles and quadrupoles, if resolved, would provide important information of sub-voxel tissue and cellular architecture. Similar to the important role that NMR has played in untangling molecular structure (Cavalli et al., 2007, Otting et al., 1990, Wishart et al., 1992), imaging higher-order frequency variation could provide a powerful tool for probing tissue microstructure such as brain connectivity noninvasively.

The backbone of brain connectivity is composed of bundled long projecting axons. Structurally, this connectivity backbone may be compared to the backbones of macromolecules. Ordered arrangement of atoms along the chain axis of macromolecules gives rise to an NMR measurable anisotropic susceptibility tensor. Similarly, on the tissue scale, the ordered arrangement of axon bundles also produces anisotropic frequency (He and Yablonskiy, 2009) and susceptibility (Lee et al., 2010, Li et al., 2012b, Liu, 2010). Although the mean susceptibility of a voxel can be measured with a gradient echo (de Rochefort et al., 2008, Li, 2001, Salomir et al., 2003), it does not measure the orientation dependence of the susceptibility (Li et al., 2011). To measure the anisotropy of magnetic susceptibility, the method of susceptibility tensor imaging (STI) has been used (Liu, 2010). A recent study also explored the capability of STI for tracking neuronal fibers in 3D in the mouse brain ex vivo (Liu et al., 2012). In large fiber bundles, the orientation determined by STI was found to be comparable to that by diffusion tensor imaging (DTI) of diffusion anisotropy (Basser et al., 1994, Basser et al., 2000, Moseley et al., 1990). However, this experimental procedure of STI requires rotating the object or the magnetic field. The requirement is clearly not convenient or even impractical for routine brain imaging on standard MRI scanners in vivo.

Here, we developed a method to measure higher-order frequency variations based on a single image acquisition without rotating the object or the magnet. This method utilized a multipole analysis of the MRI signal in a sub-voxel Fourier spectral space termed “p-space” for short. By sampling the p-space with pulsed field gradients or by shifted image reconstruction, we were able to measure a set of dipole and quadrupole susceptibility tensors. We illustrated the methodology in a simulation of aligned axons and demonstrated its use for 3D high-resolution imaging of mouse brains ex vivo at 9.4 Tesla and human brains in vivo at 3.0 Tesla. We anticipate that the p-space approach may provide a powerful method for studying tissue microstructure and brain connectivity in vivo and non-invasively.

Section snippets

The spectral space (p-space) of microscopic magnetic field

For a given imaging voxel containing heterogeneous structures, magnetic field within the voxel is also heterogeneous due to the interaction between tissue and external field. The total magnetization of the voxel is an integral of all spins within the voxel, each experiencing a slightly different local magnetic field. The phase angle of the resulting integral represents the amplitude of the mean field. The spatial heterogeneity, however, is lost during the ensemble averaging. If the field

Simulation of axon bundles

We first verified the validity of the approach using a simulated bundle of parallel axons that was situated in a cubic voxel (Fig. 2a). Without noise, both magnitude and frequency showed a quadratic relationship with p as illustrated for five representative orientations (Figs. 2b and c). The linear term was absent due to the symmetry of the phantom and the properties of Fourier transform which state that the Fourier transform of an even function is even. While the magnitudes were similar among

Discussion

At high fields, the weak interaction between magnetic fields and biological molecules is sufficiently strong to create a frequency shift in the Larmor frequency of nearby nuclear spins. Despite the paramount importance that frequency shift has attained in NMR, the utilization of frequency shift in MRI has been very limited. While measuring frequency shift and its anisotropy has enabled NMR to determine structures of large molecules, MRI has not been able to routinely utilize the vast

Acknowledgments

We thank G.A. Johnson for access to the 9.4T scanner at Duke Center for In Vivo Microscopy, Y. Qi, G. Cofer and R. Dibb for assistance in animal preparation and data acquisition. We thank A.W. Song and C. Petty of the Brain Imaging and Analysis Center for assistance with computing on the cluster.

References (33)

  • W.T. Dixon

    Simple proton spectroscopic imaging

    Radiology

    (1984)
  • J.H. Duyn et al.

    High-field MRI of brain cortical substructure based on signal phase

    Proc. Natl. Acad. Sci. U. S. A.

    (2007)
  • G.H. Glover et al.

    Three-point Dixon technique for true water/fat decomposition with B0 inhomogeneity correction

    Magn. Reson. Med.

    (1991)
  • E.M. Haacke et al.

    In vivo validation of the bold mechanism: a review of signal changes in gradient echo functional MRI in the presence of flow

    Int. J. Imaging Syst. Technol.

    (1995)
  • X. He et al.

    Biophysical mechanisms of phase contrast in gradient echo MRI

    Proc. Natl. Acad. Sci. U. S. A.

    (2009)
  • D.I. Hoult et al.

    Sensitivity and power deposition in a high-field imaging experiment

    J. Magn. Reson. Imaging

    (2000)
  • Cited by (25)

    • MRI gradient-echo phase contrast of the brain at ultra-short TE with off-resonance saturation

      2018, NeuroImage
      Citation Excerpt :

      It is generally believed that water within the myelin sheath has shorter T1, T2, T2* relaxation times and smaller diffusion coefficients. Recent theoretical analysis, simulation and in vivo experiments have also shown a range of frequency values or magnetic field perturbations within myelinated white matter (Chen et al., 2013; Cronin et al., 2017; Liu and Li, 2013; Sukstanskii and Yablonskiy, 2013; Wharton and Bowtell, 2012). For example, anisotropic magnetic susceptibility of myelin lipids has been shown to result in distinctive field patterns in the three compartments separated by the myelin sheath.

    • A method for estimating and removing streaking artifacts in quantitative susceptibility mapping

      2015, NeuroImage
      Citation Excerpt :

      There are many reasons that could contribute to this discrepancy. One reason is that macroscopic magnetic susceptibility of the white matter is anisotropic, and therefore the magnetic response of the brain cannot be adequately described using the isotropic magnetic susceptibility model (Li and Liu, 2013; Li et al., 2012a,b; Liu, 2010; Liu and Li, 2013; Liu et al., 2012a, 2013; Wisnieff et al., 2013; Xie et al., 2014). The second reason is that white matter phase is nonlinearly evolving with TE (Schweser et al., 2011a), which is also dependent on TR and flip angle (Li et al., 2013).

    • Quantitative susceptibility mapping: Current status and future directions

      2015, Magnetic Resonance Imaging
      Citation Excerpt :

      In addition to the deep gray matter structures and veins, the white matter in the brain is also of great interest. Recently, both phase and R2* relaxation were found to be dependent on the orientation of the white matter fiber tracts with respect to the main magnetic field [116–123]. This has been attributed to the myelin content, as demonstrated in a demyelinating mouse model [120].

    • Gradient echo based fiber orientation mapping using R2* and frequency difference measurements

      2013, NeuroImage
      Citation Excerpt :

      To improve upon our previous application of the HCFM, which utilised a single nerve fiber, the more realistic case of multiple fibers of different sizes was simulated. Recent studies have shown that similar multiple fiber models can produce simulations of the signal evolution that agree with in vivo measurements (Chen et al., 2013; Liu and Li, 2013; Sati et al., 2013). Hollow cylinders with outer radii of 10, 15, and 20 pixels were randomly distributed within a 1001 × 1001 matrix to yield a FVF of 0.5 within a circular region of interest (ROI) of 400 voxels in radius.

    View all citing articles on Scopus
    View full text