Semi-automated software for the three-dimensional delineation of complex vascular networks.

The understanding of tumour angiogenesis is of great importance in cancer research, as is the tumour response to vascular-targeted drugs. This paper presents software aimed at aiding these investigations and other situations where linear or dendritic structures are to be delineated from three-dimensional (3D) data sets. This software application was written to analyse the data from 3D data sets by allowing the manual and semi-automated tracking and delineation of the vascular tree, including the measurement of vessel diameter. A new algorithm, CHARM, based on a compact Hough transform and the formation of a radial map, has been used to locate vessel centres and measure diameters automatically. The robustness of this algorithm to image smoothing and noise has been investigated.