Electrocardiogram signals classification using discrete wavelet transform and support vector machine classifier

The electrocardiography allowed us to make a diagnosis of several cardiovascular diseases by representing the electrical activity heart over time; this representation is called electrocardiogram (ECG) signal. In study we have proposed model based on processing ECG signal wavelet decomposition using discrete transform (DWT). This firstly makes it possible denoise then extract statistical features from approximation coefficients denoised and finally classify data obtained in support vector machine (SVM) classifier with cross validation for more credibility. After having tested different mother wavelets at scales, accuracies fourth scale are high best accuracy 87.50%.