ABSTRACT
Time series data for heart rate (HR), blood pressure, breathing frequency and a number of other physiologic variables exhibit moment to moment variability. As a general principle, this increased fine scale variability seems characteristic of healthy, adaptable physiology (West, 2006). The variability within the data contains potential information that is lost in the traditional calculation of a mean value. Variability analysis encompasses a variety of mathematical techniques that have been applied to physiologic time series data such as: time domain analysis, frequency domain analysis, fractal analysis and entropy analysis (Seely and Macklem, 2004). Sample Entropy (SampEn), a nonlinear method for variability analysis of time series data, characterizes the inherent regularity of a data sequence (Richman and Moorman, 2000). A higher entropy score implies decreased predictability of sequential values - less self-similarity in the data.
