Cement Placement Optimization in Femoral Augmentation Using an Evolutionary Algorithm

Abstract

We used the method of Bi-directional Evolutionary Structural Optimization (BESO) to optimize the cement placement in finite element (FE) models of osteoporotic femur specimens. Two different initial conditions, i.e. no initial cement and fully cemented, were used and both converged to the same optimal cement pattern. On average, BESO predicted that, if optimized, augmentation with only 18.6ml of cement will result in 100% increase in the yield load of the models. Simulations also showed a linear relationship between the volume of the cement and the models’ stiffness and yield load. Models initially filled with cement had a much more uniform stress distribution among the cemented elements when optimized, compared to the starting configuration. Results suggest that restoring the mechanical properties of osteoporotic femurs is possible with minimal and, therefore, potentially safe volumes of cement.