Diffusion weighted images have a long tradition and are mainly used so far to identify ischemia or cell density. With the use of additional field gradients for diffusion directions and different b-values the identification of white matter tracts (diffusion tensor imaging, DTI) has become possible.

Current research projects aim at identifying different fiber directions within a single voxel and to use this information in tracking algorithms. Another topic is the development of brain segmentation algorithms based on high resolution diffusion weighted images.

A new technique, called diffusion kurtosis imaging (DKI), that evaluates the non-Gaussian diffusion properties of tissue has become available recently. It provides independent and complementary information to that acquired with traditional diffusion techniques and is believed to represent the microstructural environment of the imaged tissue.

Goal of the current work is to optimize DKI protocols for clinical routine imaging and to identify cut-off values for glioma grading.