GPU-based visualization techniques for the interactive exploration of diffusion MRI data

T.H.J.M. Peeters

Onderzoeksoutput: ScriptieDissertatie 1 (Onderzoek TU/e / Promotie TU/e)

618 Downloads (Pure)


Diffusion-weighted imaging techniques such as Diffusion Tensor Imaging (DTI) and High-Angular Resolution Diffusion Imaging (HARDI) are relatively new MRI techniques that give information about diffusion of water molecules in tissue. This diffusion gives information about the structure of underlying tissue and can help in identifying connections in the brain and the structure of the heart-muscle. It is also used to gain new insights in brain and heart diseases. Because diffusion data is high-dimensional, the visualization of such datasets is challenging. In this thesis, we present various new methods for the visualization of DTI and HARDI data. We make use of the features of modern GPUs to generate smooth and intuitive images that help to make the analysis of the data more effective and efficient. We show how to intuitively visualize the fiber orientations in a slice of data from ex-vivo scans of mouse hearts. Then we present methods for rendering glyphs to represent the tensors from DTI and probability density functions (PDF) and orientation density functions (ODF) from HARDI scans of human brains. These new visualization methods can help researchers to interactively explore DTI and HARDI data and to verify new and existing filtering methods that smoothen the data or extract higher-order structures such as fibers to show connectivity in the brain.
Originele taal-2Engels
KwalificatieDoctor in de Filosofie
Toekennende instantie
  • Department of Biomedical Engineering
  • ter Haar Romeny, Bart M. , Promotor
  • Vilanova, Anna, Co-Promotor
Datum van toekenning9 dec 2009
Plaats van publicatieEindhoven
Gedrukte ISBN's978-90-386-2066-4
StatusGepubliceerd - 2009

Vingerafdruk Duik in de onderzoeksthema's van 'GPU-based visualization techniques for the interactive exploration of diffusion MRI data'. Samen vormen ze een unieke vingerafdruk.

  • Citeer dit