Abstract
Understanding the behaviour of R-R interval data and its successive differences is critical to the dynamics of cardiac control. Several time domain measures that quantify R-R interval variability have important clinical significance in terms of risk stratification and evaluating the effectiveness of treatment procedures. The present approach at examining the distributions of successive beat-to-beat differences of R-R interval data from different populations and under different conditions (baseline and reaction times) provides a valuable insight into their previously unexplored distributional properties. In particular, our analysis reveals that the successive differences have non-normal statistical distributions (a contradiction to the commonly used assumption of normality), and the absolute successive R-R interval differences approximately follows a Weibull distribution. As an illustration of the utility of this approach, we explore the statistical properties of the time domain measure: root mean square successive difference, study the association between the Weibull scale parameter estimate and respiratory sinus arrhythmia, and propose improvements in artifact detection algorithms.
Original language | English (US) |
---|---|
Pages (from-to) | 437-451 |
Number of pages | 15 |
Journal | Statistics in Medicine |
Volume | 24 |
Issue number | 3 |
DOIs | |
State | Published - Feb 15 2005 |
Keywords
- ECG
- Heart rate variability
- Successive difference
- Weibull distribution
ASJC Scopus subject areas
- Epidemiology
- Statistics and Probability