Combination Data Mining Methods with New Medical Data to Predicting Outcome of Coronary Heart Disease

Abstract

The prediction of survival of Coronary Heart Disease (CHD) has been a challenging research problem for medical society. The goal of this paper is to develop data mining algorithms for predicting survival of CHD patients based on 1000 cases .We carry out a clinical observation and a 6-month follow up to include 1000 CHD cases. The survival information of each case is obtained via follow up. Based on the data, we employed three popular data mining algorithms to develop the prediction models using the 502 cases. We also used 10-fold cross-validation methods to measure the unbiased estimate of the three prediction models for performance comparison purposes. The results indicated that the SVM is the best predictor with 92.1 accuracy on the holdout sample artificial neural networks came out to be the second with 91.0 accuracy and the decision tress models came out to be the worst of the three with 89.6% accuracy. The comparative study of multiple prediction models for survival of CHD patients along with a 10-fold cross- validation provided us with an insight into the relative prediction ability of different data.