Diffusion MRI has established itself as an invaluable tool for the non-invasive probing of tissue microstructure and dynamics. It gave rise to unique opportunities in brain diagnostics such as stroke and demonstrated a remarkable success in assessing white matter fibre orientations. Conventional methods, however, suffer from intrinsic limitations based on a simplified assumption of Gaussian diffusion. Serious limitations are also related to the low angular resolution of the standard fibre tracking methods. We design advanced methods allowing us to essentially increase both the angular resolution (HARDI) and the range of the diffusion-encoding gradient strengths (non-Gaussian diffusion). In turn, this provides us with enhanced information about diffusion mechanisms, the underlying microstructure and brain function. In cooperation with internal and external partners we examine the potential applications of the new tools with respect to brain monitoring and diagnostics.
Last updated: 07 Mar 2014
Non-Gaussian Water Diffusion in Brain Tissue
The aim of this project is to exploit non-Gaussian water diffusion in the brain beyond the conventional range of diffusion weightings in order to develop new tools and biomarkers for monitoring various developmental and pathological changes of the tissue microstructure.
Non-Gaussian diffusion MRI in development and ageing
WM plays a vital role in information transfer between various grey matter regions.
Optimal control theory in diffusion MRI
Design of radio-frequency pulse shapes ameliorating specific features of the spin system is very well known and important problem in modern magnetic resonance imaging.
Artificial Phantoms for Studies of Anisotropic Diffusion in the Brain
Diffusion Weighted Magnetic Resonance Imaging (DW-MRI) provides access to fibre pathways and structural integrity in fibrous tissues such as white matter in the brain.
Random Walks in Model Brain Tissue
Brain tissue is often modeled as porous media. The aim of this work is to conduct random-walk Monte Carlo simulations in well-defined geometrical model systems and to establish the quantitative relationships between diffusion properties and structure
Development of Postprocessing Methods for HARDI Data Analysis
The aim of this project is to develop new reconstruction, segmentation and tractography algorithms able to infer multiple fibre structures from High Angular Resolution Diffusion Imaging (HARDI)
Noise Reduction in MR Images
Noise represents a crucial problem for any MRI technique when the signal-to-noise ratio becomes rather low. The aim of this project is to develop advanced noise correction algorithms based on the assumption of spatially inhomogeneous noise fields and using voxel-by-voxel analysis