Application of Machine Learning on ECG Signal Classification Using Morphological Features

An electrocardiogram (ECG) is a simple test that is used to check one's heart's electrical activity. Sensors attached to the skin are used to detect the electrical signal produced by one's heart each time it beats. Many people around the world suffer from cardiovascular diseases. So it is important to detect arrhythmia/abnormal and normal ECG signal more accurately. In this paper, ECG signal is classified by support vector machine (SVM) and neural network. The research is conducted on the normal and arrhythmia ECG datasets obtained from the PhysioNet website. The raw ECG data are preprocessed using different filters and then the features are extracted based on morphological values of the waveform. Twelve features are extracted and these are used to train classifiers to classify normal and abnormal ECG data. For SVM classifier, the accuracy is around 87% while for artificial neural network, MATLAB's pattern recognition app is used where the classification accuracy found is around 90% - 93%. The accuracy varies for different numbers of hidden neurons. Diverse ECG databases are used to prove the efficacy and robustness of the use of proposed morphological features and obtained results are compared with existing state-of-the-art research works. This work can be further developed in the future by incorporating deep learning for better performance and it can eventually help to detect cardiac diseases.