Constrained-Based Region Growing Using Computerized Tomography-Based Finite Element Analysis

This paper presents topological analysis of generating segmented quality meshes through constrained-based region growing. In the algorithm we first calculate the average of the markers' coordination to get the seed point. Then the novel method for calculating the difference between the neighbours and the average intensity of the seed points as well as using the modification of the Euclidean distance. Finally the method iterates between region growing and surface fitting to maximize the larger number of connected vertices approximated by a single underlying surface. These steps verify all the goals of effectiveness on real data sets from computerized tomography images.