engleski

A new method for parameter identification in a lumped model of pulmonary circulation

To obtain input impedance of pulmonary circulation, the simultaneous measurements of pulmonary artery pressure (pPA) and flow (vpv) at the root of pulmonary artery (PA) are required. For the purpose of identification of PA parameters, the pulmonary wedge pressure (pw) should be provided (usually considered as a constant). In the identification procedure it is common to take vpv and pw as input variables and to tune lumped model parameters to minimize the root mean squared error (RMSE) between measured pPA and pPAM (the pPA reconstructed from a mathematical model). Here we proposed a new method for the calculation of pulmonary input impedance and model parameters based on Swan-Ganz catheter recordings of the right ventricle pressure (pRV) and vpv obtained by echo-Doppler. Although we have readings of pPA we can not use it since it is not measured at the same location as the vpv. We hypothesize that during the ejection time pPA at the PA root may be well approximated by the pressure pPAB obtained from pRV using the Bernoulli equation. Model parameters are tuned to minimize RMSE between pPAB and pPAM during the ejection time. In addition the time varying pRV may be calculated from the tricuspid regurgitant flow, so the method shows a potential for non- invasive identification of the lumped model parameters of pulmonary circulation.

non-invasive method; six-element windkessel model; least squre method

nije evidentirano