Intelligent Prediction of Cardiac Abnormality

Cardiac abnormalities are very common in today’s world. Efficient and early prediction will help in saving the life of several individuals. ECG is a standard technique for prediction of any kind of abnormality prevailing in the heart. An intelligent cardiac abnormality prediction system is developed in this study. The study utilizes the real-time ECG data collected from a hospital. The contributing attributes are selected. The five chosen attributes include QRS value, heart rate, P-axis, R–axis, and T-axis. Machine learning techniques are explored for developing the system. Performance metrics are computed for each classifier and best performing classifier is selected for deploying the model. Naive Bayes has performed well by giving 93.50% accuracy.