Semi-analytical finite-element based method for inverse characterization of cortical bone using low-frequency guided waves

Abstract

Axial transmission research has demonstrated that low-frequency ultrasonic guided waves are sensitive to changes in the intracortical bone, which is of interest since the resorption in the endosteal region is associated to early-stage osteoporosis. Current methods rely on inversion schemes used to match experimental data with the theoretical data obtained from simplified models. However, due to the importance of the cross-sectional curvature of the cortical bone at low-frequency (e.g., <200 kHz), the implementation of a more elaborate model remains an open issue. Thus, the aim of this paper is to introduce a semi-analytical finite-element (SAFE) model to be used along with a genetic algorithm for the inverse characterization of cortical bone. Our proposal is to validate an inverse scheme using laboratory-controlled measurements on bone-mimicking phantoms at low frequency. An arbitrary cross-sectional geometry, instead of a plate or cylinder simplification, was implemented. Despite a computationally expensive SAFE routine, the results show that the model outputs estimated by the genetic algorithm are in good agreement with the reference values obtained by µCT images. The possibility of implementing parallel computation using graphics processing units in order to increase the level of complexity of the SAFE model may now be investigated.