基于分形维数的心率变异分析

心率变异性（ＨＲＶ）分析是近年来发展起来的一种新的用于无创检测心脏植物神经功能的领域．目前已提出的各种心率变异指标在对心血管疾病的早期诊断、监护和预后评估中起了不可忽视的作用，特别是对心性猝死的预报和心肌梗死的预后判别等临床应用中有重要意义．理论分析和实验结果表明，心率信号具有近似分形的性质，是非线性的，因此，用分形维数表征心率变异性是符合逻辑的．文中作者在计算了多例心率信号的分形维数并考察了相应的功率谱的基础上，提出一种新的指标——分形维数对指标．此外，作者还提出了在低采样率心电数据中精确定位Ｒ峰的新算法，并且讨论了基于纽曼－皮尔逊准则的正常心动周期的判决方法

Fractal Dimension Based Analysis of Heart Rate Variability

Analysis of heart rate variability (HRV) is a rather new field reclaimed in recent years for uninvasive inspection of cardiac autonomic nerves function. Numbers of indices proposed have attained great values in early stage diagnosis, monitoring and prognosis evaluation of cardiovascular diseases. Besides, they are of clinical significance in sudden cardiac death (SCD) prediction and myocardial infarction (MI) prognosis judgment. Theoretical analyses and experimental results indicate that heart rate signals are approximately fractal, which means that it is nonlinear. It seems logical to therefore interpret HRV via fractal dimension (FD). In this paper the authors calculate the FDs of many heart rate signals and examine their power spectra. Furthermore, a new index fractal dimension pair (FDP). Experimental results prove the fractal properties of heart rate signals and show that is proposed using FDP as the index of HRV is encouraging. Also a new method is developed for accurately positioning R peaks in ECG data sampled at a low rate and the decision of normal cardiac cycle under the Neyman Pearson Criterion is discussed as well.