A Novel Method of Vessel Segmentation for X-ray Coronary Angiography Images

This paper presents a new automatic region-growing method for vessel segmentation in two-dimensional X-ray coronary angiography images. The method consists of two parts: the feature map extraction based on a novel vesselness function; and the segmentation process which includes automatic seed-point selection, main branch segmentation and vessel detail repair. Both the greyscale and spatial information are extracted for segmentation based on region growing algorithm. The presented method is validated on several clinical X-ray coronary angiography images, and the experimental results show that the method can not only segment large vessels but also small vessels.