A genetic algorithms-based optimization method for estimating thickness and porosity of cortical bone from guided wave measurements

A step forward towards ultrasonic characterization of cortical bone has been recently reached by showing that cortical bone behaves as a waveguide for ultrasound. Measurements of the guided modes dispersion relations, along with appropriate waveguide modeling, have the potential for inferring strength-related factors such as stiffness (determined by cortical porosity) and cortical thickness. Current estimation methods rely on a number of assumptions for pairing the incomplete experimental data with the theoretical Lamb branches (e.g., initial guess on the model parameters, assignment of the data to experimental trajectories, exhaustive search, fixed number of theoretical branches, etc.). The present work introduces an efficient optimization method that strongly reduces those prior assumptions.