Coronary artery segmentation in X-Ray Angiographic image by means of a shape based level set method

Abstract

This paper presents a level set technique to extract vascular structures in X-Ray Angiographic images. It makes uses of the Chan and Vese model applied to images of non-uniform illumination and uses a shape-based model to perform the segmentation. The shape model is computed using string matching techniques. Its performance, using different metrics, has been evaluated on a image sequence of 64 angiographic images by comparison with expert delineation. A sensitivity of 81% and a specificity of 94% were found in the quantitative validation analysis.