Heart Disease Prediction Using ECG Signal Classification

Abstract

The major life threatening disease is the arrhythmia. The cardiac arrests are mainly due to arrhythmia. Most of the people can be suffered from the heart disease. The main reason is the high pressure and the food habit. In this paper they propose the heart disease detection using the image processing technique and the classification of the arrhythmia. The arrhythmia can be classified using the Discrete Wavelet transform method DWT it can be applied to the each heart beat which helps in getting dynamic and morphological features. The periodic value can be obtained by the ECG signal. The RR interval can helpful in improving arrhythmia classification. The various set of bands has been subjected for the noise reduction and the extraction of the morphological data. The neural network algorithm is applied for the classification of the arrhythmia. This method is been compared with the MIT-BIH arrhythmia database in which has 12456 beats and the ventricular arrhythmia database in which as 23456 beats. This method has the high accuracy.