Convergence of a genetic algorithm for estimating cardiac material properties

In this paper we present the results of a detailed study of a systematic numerical scheme for estimating material parameters for ventricular myocardium. The numerical scheme combines a real encoded genetic algorithm with nonlinear finite element analysis. The primary objective of this study was to determine optimal population size for the genetic algorithm so as obtain rapid convergence to actual material parameter values. Optimal parameter settings for the genetic algorithm and interdependencies of material parameters were determined through multiple optimization runs.