Observer of the Human Cardiac Sympathetic Nerve Activity Using Blind Source Separation and Genetic Algorithm Optimization

The heart rate (HR) and the arterial blood pressure (ABP) in humans can be considered under certain constraints as convoluted mixtures of cardiac sympathetic (CSNA) and parasympathetic nervous activity (CPNA). The authors present a method permitting blind reconstruction of the CSNA and CPNA using only the HR and the APB. The reconstruction model is based on blind source separation/deconvolution techniques (BSSD). Two associated learning schemes are presented and compared, a stochastic gradient and a genetic algorithm. Blind reconstruction of the CSNA using a genetic algorithm yields very good results