Classification of Cardiac Arrhythmia Using Machine Learning Algorithms

Heart is a very significant organ which aids in proper functioning of all the other organs, its preliminary function is to pump blood throughout our body and it is also considered to be the primary organ of our circulatory system. Arrhythmia is one of those common disorders which occur due to an irregular heartbeat and hence it is also known as heart rhythm problem which can be best detected with the help of Electrocardiogram (ECG). ECG is a test often conducted to record the electrical signals in the heart to determine any potential heart problems one might possess. As the Arrhythmia disease might have many adverse effects on its patients it is important to take preventive measures and categorize the patients on time and this can be accomplished with the help of various Machine Learning Algorithms such as Decision Tree, K-Nearest Neighbor, Support Vector Machine (SVM), XGBoost and Ensemble Method. Here in this paper, we are utilizing the aforementioned Machine Learning algorithms on the dataset gathered from figshare official website and consequently preprocessing the data using Python Programming language. As a result, the accuracy percentage for the correct prediction of the result for all the 5 models are compared and Ensemble Random Forest algorithm is said to have provided the best outcome with an accuracy of 98%.