An Intelligent System for Noninvasive Diagnosis of Coronary Artery Disease with EMD-TEO and BP Neural Network

A novel intelligent noninvasive diagnosis system of Coronary Artery Disease (CAD) is proposed based on Empirical Mode Decomposition (EMD)-Teager Energy Operator (TEO) and Back-Propagation (BP) neural network. The occluded arteries can produce the diastolic murmurs with high frequency energy. Firstly, the instantaneous frequency of the diastolic murmurs is estimated by EMD-TEO to identify features associated with coronary stenoses. Secondly, statistical quantities of Instantaneous frequency are extracted which are used as feature vectors relating to normal and abnormal recordings. In end, the BP neural network classifier is build to classify the extracted features. The performance of the developed system has been evaluated in clinic recordings. The correct classification rate is over 85% for normal and abnormal subjects. The results show that the intelligent noninvasive system based on EMD-TEO and BP neural network can be used as an effective tool to diagnose Coronary Artery.