Home
Tags
No Tags
Modeling of Heart Period Data And A Study of the Associated Spectral Measures
| Title: | Modeling of Heart Period Data And A Study of the Associated Spectral Measures |
| Author: | Mandrekar, Sumithra J |
| Description: | The variability around the average heart rate (heart period) known as the heart rate variability (HRV) (the heart period variability; HPV) contains information about the cyclic changes or fluctuations of heart period over time and is useful for assessing the cardiovascular autonomic regulation. Components of HRV/HPV have attracted considerable attention in several diagnostic applications in the fields of medicine and psychology and also in the modeling of the neuro-cardiovascular system. We consider different stages in the study of heart period data, namely, data acquisition (artifact identification), data processing (non-stationarity and stability of spectral measures), and data analysis (modeling and quantification of HPV measures). Firstly, we examine the distribution of successive heart period differences using four different data sets from adolescents, young adults, older women and healthy dogs. On the basis of statistical analyses, we conclude that in the absence of artifacts in the study interval, the successive differences are symmetric about zero and the absolute differences approximately follow a Weibull distribution. Using this model, we explore the statistical properties of some time domain measures of HPV, propose a semi-parametric artifact detection criterion, and establish an association between the Weibull scale parameter estimate and respiratory sinus arrhythmia (RSA), a measure of variability in heart period associated with respiration. Secondly, we model the heart period data as a mixture distribution with two normal components to capture the bimodal pattern. Finally, we assess the influence of the duration of electrocardiograph recording on the spectral measures of HRV using 5-minute data from heart failure patients at baseline and post-treatment, and 3-minute data from dogs at baseline and peak-failure. We conclude that the ratio of low frequency (LF) to the high frequency (HF) components of HRV computed from 2-minute recordings are sufficiently close to those of longer duration. This is a significant finding as smaller epoch lengths can be collected with more precision and are less susceptible to non-stationarity. Conclusions vary for the other spectral measures namely total power, low frequency and high frequency measures, for the two populations. |
| Permanent Link: |
http://rave.ohiolink.edu/etdc/view?acc_num=osu1037969947
http://hdl.handle.net/2374.OX/6535 |
| Date: | 2002 |
Files in this item
| Files | Size | Format | View |
|---|---|---|---|
|
There are no files associated with this item. |
|||