New Global and Regional Featuresfor Coronary Artery Disease Detection on 2D Echocardiography

Abstract

A computer-based of cardiac diagnosis system is needed to assist a cardiologist to evaluate the coronary artery disease quantitatively. In this paper, a computer-based global and regional diagnosis is developed to identify cardiac abnormality on 2D echocardiography. The system consists endocardium tracking, boundary modeling and segmentation, and disease detection globally and regionally.A global index was proposed to predict the left ventricular function by analyzing the shape deformation. Parameters of boundary motion such as displacement, velocity, and strain rate were applied to produce local features of each segment. Fifty-four set data was used to assess the performance of the developed system. Manual interpretation from three experts was used to validate the proposed technique. The results are very promising in recognizing the LV function globally and indicating the abnormality of each segment quantitatively.