Effect of principal component analysis on diagnosing congestive heart failure patients using heart rate records

In this study, the effects of principal component analysis (PCA) in the analysis of heart rate variability (HRV) that are used in discriminating the patients with congestive heart failure (CHF) from normal subjects are investigated. After HRV measures are obtained from 29 CHF patients and 54 normals, PCA with excluding variances of 0.0% (no PCA), 0.1%, 0.5%, 1%, 5%, 10% and 20% are applied to these measures. These measures are investigated by k-means clustering. As a result, the maximum classification accuracies are improved using PCA with excluding maximum variance of %5. In this study, maximum discrimination accuracy of 86.75% is achieved with PCA of 0.1% and ten clusters (k=10).