抄録
Changes in mean heart rate have been used to describe changes in physical or mental load. In practice, heart rates are valuable criteria for the assessment of workload levels in exercise ECG test and etc. Changes in the variability of the instantaneous heart rate (cardiac arrhythmia or HRV) are more sensitive means of measuring workload. The heart rate may be identical, but the heart rate variability frequency differs. It is, therefore, important to undertake detailed examination of the cardiac arrhythmia as a time series.
The aim of this paper is to assess the cardiac arrhythmia using practical time series of beat-to-beat variation or heart rate variability. Autoregressive model based on the stochastic theory for dynamic systems was applied to a time series of R-R intervals between two successive R-tops in the ECG to assess the heart rate variability in a quantitative way from the time series. The authors have introduced a new measure ARV defined by the ratio of residual variance for autoregressive models to the variance of the time series. Consequently, the ARV is the signal-to-noise ratio of the heart rate variation. In case of a purely random series, the ARV is reduced to zero; in case of a deterministic series, the ARV is reduced to 1.
The exercise ECG tests (continuous progressive all-out treadmill runs) on 6 college students were studied. Changes in ARV usually appeared simultaneously with the onset of chest pain during physical exercise. Five different scores of assessment of the HRV were also computed and compared. The results show that the ARV based on the autoregressive model is very effective to assess the HRV.
