Computational tools for assessing cardiovascular variability

Abstract

The analysis of heart rate variability is nowadays a common method for noninvasive evaluation of the autonomic nervous system integrity. In this work we developed an integrated and modular system — FisioSinal — capable of clinical and laboratorial evaluation of the behavior of autonomic nervous system using cardiovascular signals in humans and animal models. The computational tools that were included seek to cover the currently most validated methodologies: stochastic analysis, descriptive statistics, auto-regression, fast Fourier transform, discrete Wavelets transform and baroreflex sensitivity index. Due to its limitations, we developed and validated a new analytical tool based on Hilbert-Huang transform. FisioSinal was validated and tested by analyzing a synthesized signal. The clinical applicability was demonstrated through an analysis of the cardiovascular signals of a normal individual and a patient with paroxysmal atrial fibrillation, undergoing autonomic a provocation maneuver.