Key result
Iterated unscented Kalman filter provides reasonable accuracy for identifying mock systemic arterial windkessel model parameters.
An iterated unscented Kalman filter algorithm can accurately estimate parameters for four-element windkessel models in mock circulatory loops, which may improve in vitro testing of left ventricular assist devices.
May enhance mock loop fidelity for LVAD testing; leaves open in vivo validation.
Mock circulatory loops (MCLs) have been widely used to test left ventricular assist devices. The hydraulic properties of the mock systemic arterial system are usually described by two alternative four-element windkessel (W4) models. Compared with three-element windkessel model, their parameters, especially the inertial term, are much more difficult to estimate. In this paper, an estimator based on the iterated unscented Kalman filter (IUKF) algorithm is proposed to identify model parameters. Identifiability of these parameters for different measurements is described. Performance of the estimator for different model structures is first evaluated using numerical simulation data contaminated with artificial noise. An MCL is developed to test the proposed algorithm. Parameter estimates for different models are compared with the calculated values derived from the mechanical and hydraulic properties of the MCL to validate model structures. In conclusion, the W4 model with an inertance and an aortic characteristic resistance arranged in series is proposed to represent the mock systemic arterial system. Once model structure is appropriately selected, IUKF can provide reasonable estimation accuracy in a limited time and may be helpful for future clinical applications.
No takes yet. Share an insight, caveat, or question.
Huang et al. (2011) studied Mock systemic arterial system modeling. Iterated unscented Kalman filter (IUKF) algorithm vs. Calculated values derived from mechanical and hydraulic properties was evaluated on Parameter estimates for different models. An iterated unscented Kalman filter algorithm provided reasonable estimation accuracy for identifying parameters of a four-element windkessel model representing a mock systemic arterial system.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: