Classification of Cardiac Arrhythmia Using Machine Learning Algorithms

Irregular and unexpected slow or fast heartbeat is called arrhythmia. The two lower chambers of the heart, known as the ventricles, are where an irregular heart rhythm called a ventricular arrhythmia (VA) originates same in the case of atrial fibrillation which occurs in the two upper chambers called as aorta. An arrhythmia is a dangerous medical condition that needs prompt attention. An electrocardiogram or monitor is typically needed for its diagnosis in order to capture the irregular heart rhythms. If arrhythmia is the cause of cardiac arrest, the electrocardiogram (ECG) can help identify the cause. In this research, to investigate and evaluate a few pre-processing techniques for ECG signals. As the initial stage of pre-processing, this involves a variety of techniques to eliminate noise from the ECG signals and any undesired disturbances like Equi-ripple notch filter and Kaiser Window function. Next, several ways were investigated to turn ECG signals into pictures, segments, or waveforms that can be used as input for various machine learning algorithms like Logistic regression, Random Forest, Decision Tree and KNN classifier to analyze and forecast the likelihood of ventricular arrhythmia. Finally, a comparative analysis is done to compare all the machine learning techniques using multiple pre-processing techniques.