Estimation of blood pump parameters for cardiovascular system identification

This paper describes the use of signal processing techniques to estimate the model parameters of a left ventricular assist device (LVAD). The model consisted of lumped resistance, capacitance, and inductance elements with one time-varying capacitor to estimate the cyclical pressure generation of the device using a volume signal from the device. The model parameters were estimated by least squares fit to experimental data obtained in the laboratory. The purpose of this research is to estimate the pressure and flow signals, which are usually measured through invasive physiologic sensors, for an on-line estimator to identify cardiovascular parameters of patients who are under LVAD assist. The success of this development would provide a useful tool to monitor the cardiac function of LVAD patients without indwelling sensors. A computer simulation of the pump with a cardiovascular model was developed to demonstrate the interaction between the LVAD and the cardiovascular system. The simulation results showed agreement with those from an animal experiment and thus the simulation waveforms can be used for testing the cardiovascular estimator.