Multistage Classification for Cardiovascular Disease Risk Prediction

Cardiovascular diseases are the prominent causes of death each year. Data mining is an emerging area which has numerous applications specifically in healthcare. Our work suggests a system for predicting the risk of a cardiovascular disease using data mining techniques and is based on the ECG tests. It further recommends nearby relevant hospitals based on the prediction. We propose a multistage classification algorithm in which the first stage is used to classify normal and abnormal ECG beats and the next stage is used to refine the prediction done by the first stage by reducing the number of false negatives. In this work experiments have been conducted on the MIT-BIH Arrhythmia dataset which is a benchmark dataset. The results of the experiments show that the proposed technique is very promising.