Myocardial transversely isotropic material parameter estimation from in-silico measurements based on reduced-order unscented Kalman filter

J. Xi (1, 2) , P. Lamata (1, 2) , J. Lee (2) , Philippe Moireau (3) , Dominique Chapelle (3) , Nic Smith (1, 2)

1 OUCL - Computing Laboratory
2 Biomedical Engineering Department
3 MACS - Modeling, analysis and control in computational structural dynamics

Parameter estimation from non-invasive measurements is a crucial step in patient-specific cardiac modeling. It also has the potential to provide sig- nificant assistance in the clinical diagnosis of cardiac diseases through the quantification of myocardial material heterogeneity. In this paper, we for- mulate a novel Reduced-order Unscented Kalman Filter (rUKF) applied to the left ventricular (LV) nonlinear mechanical model based on cubic-Hermite finite elements. Material parameters in the widely-employed transversely isotropic Guccione's constitutive law are successfully identified for both ho- mogeneous and heterogeneous cases. We conclude that the four parameters in Guccione's law can be uniquely and correctly determined in silico from noisy displacement measurements of material points located on the myocar- dial surfaces. The future application of this novel and effective approach to real clinical measurements is thus promising.

Domaines: Analyse numérique [math.NA]

J. Xi, P. Lamata, J. Lee, Philippe Moireau, Dominique Chapelle, et al.. Myocardial transversely isotropic material parameter estimation from in-silico measurements based on reduced-order unscented Kalman filter. Journal of the mechanical behavior of biomedical materials, 2011, 4 (7), pp.1090-1102. ⟨10.1016/j.jmbbm.2011.03.018⟩. ⟨hal-00654545⟩