The distribution of low-variability periods in human heartbeat dynamics

We study the long-term dynamics of the short-time variability level of human heart rate, an aspect which is not addressed by the traditional methods of non-linear time-series analysis. The length-distribution of low-variability periods in human heartbeat dynamics typically follows a multi-scaling power law. The values of the scaling exponents are personal characteristics and depend on the daily habits of the subjects. Though, the distribution function of the low-variability periods as a whole discriminates efficiently between several heart pathologies.