Jos Roerdink - Visualization of brain connectivity from Diffusion Tensor Imaging data

In this talk, I will discuss some recent work on the visualization and analysis of brain connectivity from Diffusion Tensor Imaging (DTI) data.

A new illustrative fiber tract visualization technique for DTI data is presented, which is inspired by hand-drawn illustration in medical textbooks. The method can also be applied using direct touch interaction on modern wall or table displays.  Next, a novel tensor interpolation method is introduced that allows DTI streamlining to overcome low-anisotropy regions. Then I discuss the application of DTI for studying brain connectivity in patients with tinnitus. Using a probabilistic DTI approach, we determined the characteristics of fiber tracts from the inferior colliculus to the medial geniculate body up to the primary auditory cortex, and also the connections between the auditory system and the amygdala, which may be involved in some forms of tinnitus.  Finally, I present a force-directed graph layout (FDGL) method as an explorative tool for connectivity-based brain parcellation studies.  It is shown how the method can be used to investigate changes in connectivity patterns in the supplementary motor area (SMA).