Vessel bifurcation detection in scale space

Several methods have been proposed for segmentation of vessels, many based on scale-space. However, none of the existing methods for blood vessel segmentation is appropriate for extension to bifurcation detection. Existing bifurcation detection algorithms use an inherently serial “track and detect” approach, requiring a seed point. We present a comprehensive scale space analysis of vascular bifurcations, resulting in a simple, novel algorithm for direct detection of blood vessel bifurcation points based not only on spatial variation across scales, but also on the variation at a single spatial point across scales, without training data or seed points. We present an analytical model for the bifurcation evolution with scale, combined with eigenvalue analysis to create a "bifurcationness" filter. We reveal, for the first time, a hybrid structure of bifurcations in scale-space. The algorithm was tested for validation in both 2D and 3D, with synthetic data and medical images.