You are here
Modelling diffusion imaging data using localized kernels
Thursday, 23 January, 2014 - 11:30 to Monday, 27 January, 2014 - 12:30
Modelling diffusion imaging data using localized kernels on the sphere
Stefan van der Walt
Magnetic resonance image (MRI) allows the observation of protons in water molecules that traverse the human brain. Mapping their motion illuminates fiber pathways, and allows us to explore the structure of the brain *in-vivo* ('within the living'). In this talk, we describe why mathematical modelling of MRI signals is needed, and present a new model based on localized kernels that fits the data more robustly than equivalent methods, while having highly desirable mathematical properties.
Stéfan is a lecturer in applied mathematics at Stellenbosch University. Last year, he completed a post doctoral fellowship at the Helen-Wills Neuroscience Institute of the University of California, Berkeley. A long time contributor to NumPy and SciPy, he now leads the development of scikit-image, the image processing toolbox for Python. Stéfan is a strong advocate for the use of open source software in science and education. In his spare time, he enjoys running and photography in the great outdoors.
Conference Room (Basement), Mattarello