High-level fusion of multi-modal brain-imaging data
Theme: Medical Visualisation
People: Dr Julien Milles (LUMC), Dr Charl P. Botha (TUD), Ir. Jorik Blaas, Ir. Frits H. Post
Description
MRI, or Magnetic Resonance Imaging, is a flexible medical imaging modality that can also be used to acquire, among others, functional MRI (fMRI) and Diffusion Tensor Imaging (DTI) data. While MRI is often employed for the study of anatomical structure, fMRI shows regions of metabolic activity in the brain and DTI shows the actual connective architecture, or more simply put the "wiring" connecting together the various computing centres in the human brain, all of this in vivo! Recently, more high level information about the brain is being studied by neuro-scientists, for example the connectome that aims at combining both anatomical and functional connectivity, respectively provided by DTI and fMRI, to describe and study consistent structure-function patterns in the human brain.
The availability of this data opens up a range of exciting new brain research possibilities. However, its inherent complexity brings with it a number of new challenges. All of this multi-modal data, including the higher-level information, has to be somehow combined in order to be studied more effectively. In addition, to be able to compare brains with each other, the data also has to be studied in a common reference frame, such as a brain-atlas. Interactive medical visualization techniques, extending some of our previous work in this regard, would greatly assist neuro-scientists in performing their research.
The figures below, courtesy of Jorik Blaas (TUD) and Gordon Kindlmann (Harvard) show DTI visualizations. Two of them integrate 3D glyphs - these could play an important role in this research project:
The Radiology Department of the Leiden University Medical Centre (LUMC) has state-of-the-art MRI scanning equipment and the necessary neuro-science know-how. In this M.Sc. project, you will work together with the TU Delft Medical Visualization group and the LUMC Radiology Department to develop new techniques for the interactive visualization of combined MRI, fMRI, DTI datasets and other high-level information, using high-level glyphs, fibre tracking, volume rendering and advanced interaction techniques. The purpose of these new techniques will be to facilitate neuro-science research.
More information
If you are interested in the possibilities of doing your M.Sc. project as part of this research or you would just like more information, please contact Dr Charl P. Botha at <c.p.botha@tudelft.nl> or Dr Julien Milles at <j.r.milles@lumc.nl>.