Detection and localization of coronary artery stenotic segments using image processing

摘要

Coronary Artery Disease (CAD) is one of the emerging causes of death all over the world. Coronary angiography techniques for the detection of CAD lead to various complications like artery-dissection, arrhythmia and even death. In this paper we propose an image processing method for detecting and localizing the stenosis regions in the artery. The method is applied on the Digital Imaging and Communications in Medicine (DICOM)) image of the heart for the detection of stenosis. The region of interest is considered as the coronary arteries and is segmented using vessel enhancement diffusion filter along with dilation and erosion morphology. Centerlines of the segmented arteries are extracted using a fast marching based method. Detection of the stenosis is done by estimating the vessel diameter at each centerline location. The abrupt reduction in the diameter of the arteries is a sign of stenosis. The proposed method is evaluated with a set of images and it gives 86.67% accuracy for detecting the stenosis. This result validates the performance of the proposed method.