Bart M. ter  Haar Romeny
Eindhoven University of Technology, Netherlands
and
Northeastern University, Shenyang, China
B.M.terhaarRomeny@tue.nl

Abstract

Electrophysiological, optical, opto-genetic, fMRI-, diffusion MRI and other brain imaging techniques have revealed an astoundingly well organized visual front-end. However, real understanding and generic modeling of the complex representations in the huge filter banks still offers many challenges. The multi-scale structure has inspired to (now in computer vision widely used) robust differential shift- and rotation invariant operators and keypoint detectors, and to hierarchical segmentation approaches and recognition techniques. The multi-orientation structure, recognized in the cortical pinwheels and their interconnections, has inspired to robust contextual tracking and adaptive enhancement operations.

We will discuss an innovative Lie-group based model for simultaneous analysis in the multi-scale, multi-orientation, multi-velocity, multi-disparity and multi-color domain. Applications will be presented for contextual, crossing preserving enhancement of elongated structures, such as 2D and 3D brain vasculature (e.g. quantitative retinal and extra-orbital vessel analysis exploited in a large-scale program for screening for early diabetes), and complex 3D brain dwMRI tractography, and perceptual grouping. The results are highly promising, and regularly outperform classical approaches, but need substantial processing, which today can be directed to, also brain-inspired, massively parallel GPU processing.

Short Bio

Bart M. ter Haar Romeny received the MSc degree in Applied Physics from Delft University of Technology in 1978, Ph.D. from Utrecht University in 1983 in biophysics. He became principal physicist of the Utrecht University Hospital Radiology Department. He was co-founder and associate professor at the Image Sciences Institute (ISI) of Utrecht University (1989-2001). From 2001, ter Haar Romeny holds the chair of Biomedical Image Analysis at the Department of Biomedical Engineering of Eindhoven University of Technology and Maastricht University in the Netherlands, and since 2011 is appointed distinguished professor at Northeastern University, Shenyang, China. His research interests include quantitative medical image analysis, its physical foundations and clinical applications. His interests are in particular the mathematical modeling of the visual brain and applying this knowledge in operational computer-aided diagnosis systems. He authored an interactive tutorial book on multi-scale computer vision techniques, edited a book on non-linear diffusion theory in computer vision. He is author of over 200 refereed journal and conference papers, 12 books and book chapters, and holds 2 patents. He supervised many PhD students, of which 4 graduated cum laude. He is senior member of IEEE, and chairman of the Dutch Society for Pattern Recognition and Image Processing.