Classification of Arrhythmia Based on Machine Learning Algorithms Using ECG Signals

Arrhythmia is the high occurring cardiovascular diseases and it is more common now a days. Early detection of arrhythmia helps doctors to diagnose the disease and to treat using appropriate medicines. One of the widely used method to find arrhythmia is by using Electrocardiogram (ECG) signals. Automatic identification and classification of arrhythmia can improve the efficiency in the diagnosis process and improve the quality of treatment. Many machine learning models and deep learning models were used by the researchers for the arrhythmia classification. This paper presents the comparison of different sampling techniques, machine learning models and ensemble models for the classification of arrhythmia. Based on the results the high accuracy of 91% is obtained by radial basis function kernel in SVM using upsampling of data.