Detection of Abnormal electrocardiograms Based on Various Feature Extraction methods

Abstract

Electrocardiogram (ECG) is a graphical representation of the electrical activity of the heart obtained by placing various electrodes on specific areas of the subject's body surface. Abnormalities in a patient's ECG signal may indicate cardiac diseases that require immediate medical attention. As a result, detecting an abnormal ECG is critical for the patient's benefit. This work develops a method for classifying ECG signals as normal or abnormal. In this paper, we propose a method for detecting cardiac arrhythmias in electrocardiograms (ECG). In the first stage, the proposal focuses on various feature extractor methods. The Fourier Transform (DFT), Discrete Wavelet Transform (DWT), and Improved complete ensemble empirical mode decomposition with adaptive noise were the feature extraction techniques evaluated (ICEEMDAN). The PCA method is then used to reduce the number of features. Finally, for classification, the Support Vector Machine (SVM) was used, which was trained using the features extracted in the first stage. The proposed models are tested using datasets from MIT-BIH arrhythmia and PTB Diagnostics. The experimental results show that using 3-PCs with the DWT method produces better results than the other methods, which achieve 98.7% in terms of accuracy.