An Automated Method for Segmentation of Coronary Arteries in Coronary CT Imaging
Yin Wang, Panos Liatsis · 2010
Many applications in the field of medical image processing require precise estimation of diagnostic parameters of the anatomical structures to be examined. In this paper, we present a novel two-step three dimensional region statistics based active contours algorithm for automatic segmentation of the entire coronary arterial trees in 3D coronary CT images. In the first phase, we define the optimal binary labels for extracting the entire tree structure of the coronary arteries from the input images. This is accomplished by a region statistics based active contours method. The possible outliers, such as the touching vessel like structures, are removed from the segmentation in the following stage by the proposed frame by frame correction algorithm.