Gap Connection Of Vascular Branches By Nonlinear Least Squares Curve Fitting Algorithm

The integrity of vascular structure plays an important role in medical diagnosis. Yet segmented vascular usually contain undesirable gaps. In order to recover the topology of the real vessel network, we need to connect the gaps between the closest discontinuous branches. A new method is proposed which merges discontinuities in three-dimensional(3D) images of vascular structures. This algorithm is based on the skeletonization of the segmented network followed by projection. And a nonlinear least square curve fitting algorithm is then applied for connection disconnections in broken branches. Curve fitting is an essential tool for analyzing biological data, while a nonlinear least square curve fitting has its advantages in dealing with the connectivity and smoothness of the whole piece of curves. The fitting method permits to merge the most common kinds of discontinuities found in vascular networks. The experimental results demonstrate the method is effective for fitting gaps of vascular structure.