Diffusion Imaging

Diffusion imaging captures the movement of water molecules, termed diffusion. Diffusion in tissue is hindered by cell membranes. Therefore, the orientation-dependent diffusion profiles provide information about tissue microstructure. For instance, when water molecules are observed in myelinated neuronal fibers, their diffusion is less hampered along than across fiber tracts. Diffusion within a voxel (i.e., a three-dimensional data point) is often captured by a tensor (i.e., ellipsoid) model. For each voxel, a tensor is fitted to the diffusion data. This method, termed diffusion tensor imaging (DTI), yields two independent scalar measures, both derived from the three eigenvalues of the tensor:

  • the mean diffusivity, which is defined as the average of all three eigenvalues, and
  • fractional anisotropy, a measure of the relative lengthiness of the ellipsoid.
Diffusion imaging | Center for Lifespan Psychology
© MPI fuer Bildungsforschung

DTI provides information about the spatial orientation of the main diffusion direction and is sometimes used as the basis for tractography, which seeks to reconstruct the fiber connections of the brain based on local orientation information. However, by providing only one directional description per voxel, DTI provides an impoverished, and at times inaccurate, picture of histological reality. For instance, the crossing of fibers may go unnoticed. To enhance the microstructural veridicality of diffusion imaging, the project is working on methods that combine information derived from diffusion imaging with information derived from two other imaging modalities, magnetization transfer saturation imaging and short-T2 relaxation-time mapping. The project is also working on Q-ball imaging, a method suited to model more complex diffusion profiles. The combined use of these data acquisition and data analysis protocols will provide a more realistic picture of age changes and age differences in the brain's structural connectivity.

Color-coded Mapping

Farbcodiertes MRI | Color-coded DTI
© MPI fuer Bildungsforschung

Color-coded mapping of principal diffusion directions: Red indicates a left–right orientation, blue is for inferior–superior, and green for an anterior–posterior orientation.

Key References

Bodammer, N. C., Kaufmann, J., Kanowski, M., & Tempelmann, C. (2004). Eddy current correction in diffusion-weighted imaging using pairs of images acquired with opposite diffusion gradient polarity. Magnetic Resonance in Medicine, 51, 188–193. doi: 10.1002/mrm.10690

Bodammer, N. C., Kaufmann, J., Kanowski, M., & Tempelmann, C. (2009). Monte Carlo-based diffusion tensor tractography with a geometrically corrected voxel-centre connecting method. Physics in Medicine and Biology, 54, 1009–1033. doi: 10.1088/ 0031-9155/54/4/013