A Monte Carlo Sensitivity Analysis for a Dimensionally Reduced-Order Model of the Aortic Dissection.

Abstract

Purpose: Aortic dissection is associated with a high mortality rate. Although computational approaches have shed light on many aspects of the disease, a sensitivity analysis is required to determine the significance of different factors. Because of its complex geometry and high computational expense, the three-dimensional (3D) fluid-structure interaction (FSI) simulation is not a suitable approach for sensitivity analysis.

Methods: We performed a Monte Carlo simulation (MCS) to investigate the sensitivity of hemodynamic quantities to the lumped parameters of our zero-dimensional (0D) model with numerically calculated lumped parameters. We performed local and global analyses on the effect of the model parameters on important hemodynamic quantities.

Results: The MCS showed that a larger lumped resistance value for the false lumen and the tears result in a higher retrograde flow rate in the false lumen (the coefficient of variation, $c_{\text{v},i}=0.0183$ , the sensitivity $S_{{\text{X}}_i}^{\sigma }=0.54$ , Spearman's coefficient, ${\rho }_{\text{s}}=0.464$ ). For the intraluminal pressure, our results show a significant role in the resistance and inertance of the true lumen (the coefficient of variation, $c_{\text{v},i}=0.0640$ , the sensitivity $S_{{\text{X}}_i}^{\sigma }=0.85$ , and Spearman's coefficient, ${\rho }_{\text{s}}=0.855$ for the inertance of the true lumen).

Conclusion: This study highlights the necessity of comparing the results of the local and global sensitivity analyses to understand the significance of multiple lumped parameters. Because of the efficiency of the method, our approach is potentially useful to investigate and analyze medical planning.