Synthetic heart rate variability signals designed to validate heart rate variability analysis

GALLARDO, J; NICOLA SIRI, L

SALTA

Congreso; XXX Congreso Nacional de Cardiología; 2012

FEDERACION ARGENTINA DE CARDIOLOGIA

In the present work, sinthetic time series which simulate the known characteristics of heart rate variability were computer - generated, to validate new methods used in heart rate variability analysis. Heart rate variability is often used as a window to explore the autonomic nervous system itself. Heart rate variability is described by means of the time series of consecutive RR intervals represented against the number of the k-beat (cardiotachogram) or against the time tk of occurrence of the corresponding k-beat (discrete events series), and can be analyzed in several ways: i) in the ?time domain? performing descriptive statistics on the RRk ensamble), ii) in the ?frequency domain? co mputing the spectrocardiogram (power spectral density vs. frequency), and iii) in the ?phase space domain? which explore the nonlinear (chaotic) dynamics embedded in autonomic control of heart rhythm. Heart rate variability analysis are in continuous revision, because of less conventional ways to perform frequencial and non linear dynamics analysis (i.e. time - frequency analysis, Fourier analysis on unevenly spaced time series, fractal geometry and entropy analysis), and the evolution of discrete time series analysis associated to faster and cheaper computer facilities. To study the efficacy and robustess of such new methods, the use of simulated cardiotachograms which reproduce the physiology and the physiopatology of heart rate variability, seems to be a powerful tool, because permits to evaluate the capability of the method under test of recovering information previously embedded in the cardiotachograms. In the present work, cardiotachograms were generated by programming in Matlab® .The esential characteristic of the heart rate variability is the fact that the ordinate in the discrete events time series (the RR duration) is also the interval which separates each point from the precedent in the abscisa, producing a particular set of unevely spaced discrete time series; this property was carefully preserved in the simulations. Different class of modulation of the time series, reproducing physiological mechanisms, were introduced in the simulated cardiotachograms. To facilitate their use for any researcher, the simulated cardiotachograms will be uploaded into the public database. Nevertheless the use of those cardiotachograms in validation of a variety of heart rate variability analysis methods, will be matter of a future work by our team, in this presentation a comparison between embedeed information and recovered information were performed, as a quality control of the simulated time series.