Cardiac Arrhythmia Detection using Ensemble Machine Learning Techniques

Arrhythmia is a sudden, irregular, slow, or rapid heartbeat. An arrhythmia is a serious medical disease that requires immediate care. In order to record the abnormal heart beats, an electrocardiogram or monitor is often required for its diagnosis. Electrocardiogram (ECG) results can assist determine whether arrhythmia is to blame for cardiac arrest. In this research we will focus on passing the ECG dataset into different machine learning algorithms like Logistic regression, Decision Tree, KNN, Random Forest classifier, Support Vector Machine and Gradient Boosting classifier. The best classifier’s hyperparameters will be fine-tuned via randomised search cross validation. All the performance measures will be recorded and the top five classifiers i.e. Decision tree and hyperparameter tuned Random Forest, Gradient boosting classifier, KNN and SVM classifier will be passed through ensemble classifier method to get a more accurate result i.e. an accuracy of 98.49%.