Automated Detection of Myocardial Ischemia Using Statistical Signal Processing and Neural Networks

Abnormal changes in the ST segment of an electrocardiogram (ECG) are very important diagnostic parameters for detecting myocardial ischemia. Myocardial Ischemia is one of the most common causes of death in the world and early diagnosis and treatment of it has a critical importance. In this study, we propose a novel method for the automatic detection of myocardial ischemic events from the electrocardiogram (ECG) signal. The ECG analysis algorithm developed in this study consists of the Discrete Wavelet Transform (DWT), statistical signal processing (PCA & ICA) and Artificial Neural Networks (ANN). The ST-T segment is obtained based on the detection of R peak location, based on the well-known Pan & Tompkins method. A neural network based system was designed for automatic detection of myocardial ischemia. The preprocessing stage (feature extraction) comprises Wavelet transform and statistical signal processing techniques (Principal Component Analysis and Independent Component Analysis) aiming at extracting discriminant information and reduce redundancy in the set of features. Through the proposed system, classification efficiencies of ~99% were achieved and the misclassification of signatures was almost eliminated

Keywords:

Independent Component Analysis , Myocardial Ischemia , Neural Networks , Principle Component Analysis , Wavelet Transform