Elsevier

NeuroImage

Volume 34, Issue 1, 1 January 2007, Pages 144-155
NeuroImage

Probabilistic diffusion tractography with multiple fibre orientations: What can we gain?

https://doi.org/10.1016/j.neuroimage.2006.09.018Get rights and content

Abstract

We present a direct extension of probabilistic diffusion tractography to the case of multiple fibre orientations. Using automatic relevance determination, we are able to perform online selection of the number of fibre orientations supported by the data at each voxel, simplifying the problem of tracking in a multi-orientation field. We then apply the identical probabilistic algorithm to tractography in the multi- and single-fibre cases in a number of example systems which have previously been tracked successfully or unsuccessfully with single-fibre tractography. We show that multi-fibre tractography offers significant advantages in sensitivity when tracking non-dominant fibre populations, but does not dramatically change tractography results for the dominant pathways.

Section snippets

Signal model

We use the model described in Behrens et al. (2003b) and Hosey et al. (2005). It is a partial volume model, where the diffusion-weighted MR signal is split into an infinitely anisotropic component for each fibre orientation, and a single isotropic component. Unlike Behrens et al. (2003b), here, we will infer on multiple fibre orientations.

The predicted signal for each diffusion-weighted measurement at each voxel is:Si=S0((1j=1Nfj)exp(bid)+j=1Nfjexp(bidriTRjARjTri))where S0 is the

Probabilistic tractography in a multi-fibre field

We perform tractography using the same sampling scheme that was shown to sample from the global probability of connection in Behrens et al. (2003b), and has also been used in Parker and Alexander (2003), Hagmann et al. (2003), Jones and Pierpaoli (2005), and Lazar and Alexander (2005).

This scheme amounts to streamline tractography except, at each step, instead of progressing along the most likely principal diffusion direction, we draw a sample from the posterior distribution on principal

60 direction data

Diffusion-weighted data were acquired on a 1.5 T Siemens Sonata scanner by using echo planar imaging (72 × 2 mm thick axial slices, matrix size 128 × 104, field of view 256 × 208 mm2, giving a voxel size of 2 × 2 × 2 mm). The diffusion weighting was isotropically distributed along 60 directions by using a b-value of 1000 s mm 2, allowing an echo time of 89 ms and a repetition time of 9 s, For each set of diffusion-weighted data, 5 volumes with no diffusion weighting were acquired at points throughout the

Local fitting of the complex fibre model

Fig. 2 shows typical examples of the probabilistic multi-fibre fit to the 60 direction data set. Note the wide extent of complex fibre architecture in Fig. 2(a). In this case, the diffusion data at one third of voxels with Fractional Anisotropy > 0.1 were able to support more than one fibre orientation, but no single voxel in the data set supported more than two orientations. In the 12 direction set, no single voxel supported more than a single fibre orientation. Note that the method is general

Discussion

We have presented a direct extension of Behrens et al. (2003b) to the case of complex fibre architecture, and shown that, using this modelling approach, it is possible to detect white matter regions of complex fibre architecture that have previously been identified by model-free inversion approaches (Tuch et al., 2003, Tuch et al., 2005, Jansons and Alexander, 2003, Tournier et al., 2004, Ozarslan et al., 2005). Furthermore, by inferring on a model of local diffusion, we are able to limit the

Acknowledgments

The authors would like to acknowledge funding from the UK Medical Research Council (TEJB, HJB), the Royal Society (MFSR), the UK Engineering and Physical Sciences Research Council (MWW), the Dr. Hadwen Trust (SJ) and the Wellcome Trust (HJB). We would also like to thank Professor Paul Matthews and the Isle of Islay for providing their combined support for this work.

References (41)

  • Stephen M. Smith et al.

    Accurate, robust, and automated longitudinal and cross-sectional brain change analysis

    NeuroImage

    (2002)
  • B. Stieltjes et al.

    Diffusion tensor imaging and axonal tracking in the human brainstem

    NeuroImage

    (2001)
  • J.-Donald Tournier et al.

    Direct estimation of the fiber orientation density function from diffusion-weighted MRI data using spherical deconvolution

    NeuroImage

    (2004)
  • David S. Tuch et al.

    Diffusion MRI of complex neural architecture

    Neuron

    (2003)
  • Daniel C. Alexander

    Multiple-fiber reconstruction algorithms for diffusion MRI

    Ann. N. Y. Acad. Sci.

    (2005)
  • Yaniv Assaf et al.

    New modeling and experimental framework to characterize hindered and restricted water diffusion in brain white matter

    Magn. Reson. Med.

    (2004)
  • T.E.J. Behrens et al.

    Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging

    Nat. Neurosci.

    (2003)
  • T.E.J. Behrens et al.

    Characterization and propagation of uncertainty in diffusion-weighted MR imaging

    Magn. Reson. Med.

    (2003)
  • J.M. Bernardo et al.

    Bayesian Theory

    (2000)
  • Paula L. Croxson et al.

    Quantitative investigation of connections of the prefrontal cortex in the human and macaque using probabilistic diffusion tractography

    J. Neurosci.

    (2005)
  • Cited by (2840)

    View all citing articles on Scopus
    View full text