Statistical Analysis of Heart Rate Data

Abstract

This talk includes two parts of our work in analyzing heart rate data. In the first part, a quadratic classification rule is applied to extract several important ECG diagnosis-aiding indices among normal children and children with ventricular septal defect with or without congestive heart failure. Best classification vectors are searched for pairwise classification. Two methods, minimum distance criterion and a two stage classification procedure, are considered for three way classification. Logistic regression models based on transformations of these important diagnosis-aiding indices are also proposed. The receiver operating characteristic curves of the proposed models show better performance than those of linear and quadratic logistic models. In the second part, spectral analysis of heart rate variability were applied to investigate autonomic nervous system activity. Two important risk factors related to heart rate variability are considered. The first risk factor, a useful measure of sympathetic/parasympathetic balance, is the ratio of low frequency to high frequency spectrum power. Since the distribution of the conventional ratio statistic are unknown, in our study we propose and study an equivalently useful ratio statistic. The second risk factor is related to low variability of heart rates. We establish control charts monitoring the ratio risk factor and low heart rate variability for patients after operation in vascular intensive care units.