Mathematical morphology analysis of 3D MRA images of human brain for estimation of blood vessels parameters

The aim of the described research is to develop brain blood vessels modeling algorithm on the basis of three-dimensional high-resolution magnetic resonance images. The geometrical model of vessels will be used to visualize arteries and veins and to determine their quantitative description. It is expected that the estimated geometrical parameters of vessels will significantly complement the results of medical examinations and contribute to a more objective and accurate medical diagnosis. Geometrical parameters need to be extracted from MRA image after segmentation. Most important parameters are: vessel diameter, local direction and location of vessel endings and bifurcations. In this paper, the estimation algorithms will be presented. They are developed on the basis of three-dimensional skeleton of vessels and vessel tracking techniques.