Tractography: where do we go from here? - PubMed (original) (raw)

Review

Tractography: where do we go from here?

Saad Jbabdi et al. Brain Connect. 2011.

Abstract

Diffusion tractography offers enormous potential for the study of human brain anatomy. However, as a method to study brain connectivity, tractography suffers from limitations, as it is indirect, inaccurate, and difficult to quantify. Despite these limitations, appropriate use of tractography can be a powerful means to address certain questions. In addition, while some of tractography's limitations are fundamental, others could be alleviated by methodological and technological advances. This article provides an overview of diffusion magnetic resonance tractography methods with a focus on how future advances might address challenges in measuring brain connectivity. Parts of this review are somewhat provocative, in the hope that they may trigger discussions possibly lacking in a field where the apparent simplicity of the methods (compared to their functional magnetic resonance imaging counterparts) can hide some fundamental issues that ultimately hinder the interpretation of findings, and cast doubt as to what tractography can really teach us about human brain anatomy.

PubMed Disclaimer

Conflict of interest statement

Author Disclosure Statement No competing financial interests exist.

Figures

FIG. 1.

FIG. 1.

Cartoon illustrations of ambiguities in mapping diffusion to axon geometry, and their consequences for tractography. Top left: Different axon geometries can lead to a similarly oriented tensor. The Tensor's principal direction is the same for all cases, but modeling crossing fibers helps distinguish a few of the cases. Modeling fiber fanning separates the top two geometries. Further modeling the polarity of a fanning can help separate all cases. Top right: Illustration of the asymmetry in tracking when fanning polarity is modeled locally. Bottom left: Illustration of a case of kissing fibers, where the local model is one of crossing fibers. Tractography will lead to jumping between the tracts, causing false positives. Bottom right: case of ambiguities near the cortex. Both axon configurations lead to the same diffusion profile (and hence the same tracking results), but have very different implications in terms of the actual connectivity.

FIG. 2.

FIG. 2.

Example of diffusion tensor imaging-derived images that show the extra white-matter contrast gained by capturing diffusion anisotropy. The PDD (principal diffusion direction) map on the right is colored according to Red, left-right; Green, anterior-posterior; Blue, superior-inferior. Cing, cingulum bundle; CC, corpus callosum; CR, corona radiata, SLF, superior longitudinal fasciculus.

FIG. 3.

FIG. 3.

(a) Example use of tractography to define regions of interest in the corpus callosum for quantitative measurements of FA through life-span [adapted from Lebel et al. (2010)]. (b) Tractography of the cingulum bundle allows parameterization of position along the tract. This enables better inter-subject registration for averaging and statistics [adapted from Gong et al. (2005a)].

FIG. 4.

FIG. 4.

(a) Gray matter parcellation of the thalamus (top left), cingulate (top-right), medial prefrontal (top-right), Broca's area (bottom left), and lateral premotor cortex (bottom right). (b) Example use of tractography and cortical parcellation of the subgenual cingulate for guiding surgical interventions. Black dots correspond to locations where electrical stimulations were effective. Figures adapted from (a) (Anwander et al., ; Beckmann et al., ; Behrens et al., ; Johansen-Berg et al., ; Tomassini et al., 2007) and (b) (Johansen-Berg et al., 2008).

FIG. 5.

FIG. 5.

Comparison of a T1-weighted contrast (left) to the results of the super-resolution technique at 250 μm isotropic (right). Details of white matter architecture can be seen with the naked eye.

FIG. 6.

FIG. 6.

Comparison between tractography-based and resting-state fMRI connectivity (correlation of time series) thresholded at arbitrary levels. The seed point is indicated by the black cross on the left temporo-parietal junction. Tractography data are from a single subject and fMRI data from averaging correlation matrices from five subjects. Notice the similarities but also the differences between the two maps. fMRI, functional magnetic resonance imaging.

References

    1. Aboitiz F, et al. Fiber composition of the human corpus callosum. Brain Res. 1992;598:143–153. -PubMed
    1. Aganj I, et al. Reconstruction of the orientation distribution function in single- and multiple-shell q-ball imaging within constant solid angle. Magn Reson Med. 2010;64:554–566. -PMC -PubMed
    1. Aganj I, et al. A Hough transform global probabilistic approach to multiple-subject diffusion MRI tractography. Med Image Anal. 2011;15:414–425. -PMC -PubMed
    1. Alexander DC. A general framework for experiment design in diffusion MRI and its application in measuring direct tissue-microstructure features. Magn Reson Med. 2008;60:439–448. -PubMed
    1. Alexander DC. Barker GJ. Optimal imaging parameters for fiber-orientation estimation in diffusion MRI. Neuroimage. 2005;27:357–367. -PubMed

Publication types

MeSH terms

LinkOut - more resources