Predicting Coronary Artery Disease from Heart Rate Variability Using Classification and Statistical Analysis

HRV (heart rate variability) is one of the most promising quantitative indications of autonomic activity. In present study, our aim is to develop the multi-pararmetric feature including linear and nonlinear features of HRV. We also propose a suitable prediction model to enhance the reliability of medical examination for cardiovascular disease. This study analyzes the HRV for three recumbent positions. Interaction effect between recumbent positions and groups (Normal, Patient) was observed based on the HRV indices. We have carried out various experiments on linear and nonlinear features of HRV to evaluate classifiers. In our experiments, SVM and Bayesian classifiers outperformed the other classifiers.