Automatic Assessment of Cardiac Artery Disease by Using DCAD Module

Abstract

In patients with cardiac artery disease myocardial perfusion scan, which is a non-invasive method, is utilized. This study is conducted to achieve an advantageous software applicable to quantitative myocardial SPECT perfusion. Each cross-section of the left ventricle is segmented by applying fuzzy clustering method. After obtaining the myocardial skeleton of the left ventricle from its short axis cross sections, we make use of fuzzy logic to decide whether the pixel belongs to myocardial muscle and any perfusion perturbation or not. Abnormal critical conditions in rest and stress studies and coronary artery disease diagnosis were investigated in a set of about 200 images. Measurement and allocation of different myocardial sectors to specific coronary arteries were accomplished by utilizing collected information about respectively 75 patient men and 62 patient women, and the validity of artery obstruction diagnosis has been proved in 40 patients having been under the coronary angiography. We named the achieved software DCAD(b) which has demonstrated a considerably good performance in coronary artery occlusion diagnosis and would be a promising method aiding nuclear medicine specialists in diagnosing these defections.