摘要
功率谱分析是目前心率变异性分析的主要方法。文中对比周期图法简要介绍了自回归模型(Autoregressive models)功率谱估计的原理 ,并由此出发对心率变异信号进行功率谱分析。心率变异信号功率谱可划分为代表不同生理意义的特征谱段。医学上通常采用这些谱段中的所谓低频谱段与高频谱段的比值 L F/HF作为衡量心脏生理功能的一项有效指标。文中探讨了根据当前心率变异功率谱特征谱段划分方案得出的 L F/HF指标的有效性 ,提出了不同的特征谱段划分方案 ,并且从统计学的观点对由该方案得出的 LF/HF指标进行了论证 ,最后证明它的有效性及显著性均优于当前的划分方案得出的 L F/HF指标。
Power spectrum density (PSD) analysis is a popular method in heart rate variability (HRV) analysis. This paper gives a general introduction to PSD analysis with autoregressive (AR) models in contract with periodogram. It focuses on the characteristic frequency band division of PSD of HRV. Such frequency bands include low frequency band (LF) and high frequency band (HF). LF/HF means the ratio between the total energy of LF and HF. Then the validity of the parameter of LF/HF is discussed. LF/HF is an important criterion in medical application in evaluating the physiological function of heart. It is believed that the LF parameter reflects the fluctuation of blood pressure, and the HF parameter reflects the fluctuation of breath. Based on the present method of frequency band division, we notice that the LF/HF parameter is not the most efficient one to evaluate the physiological function of heart. So some new methods of frequency band division are given. As a result, on each method of the new frequency band division, an LF/HF parameter can be got. Compared with the different LF/HF parameters, the most efficient one is picked out. At last the preferred LF/HF parameter is statistically tested based on a certain number of subjects.
出处
《数据采集与处理》
EI
CSCD
2000年第2期259-262,共4页
Journal of Data Acquisition and Processing
