Finite-element modeling of bones from CT data: sensitivity to geometry and material uncertainties

The aim of the present work was to investigate the influence that the uncertainties related to the segmentation process and to the material assignment procedures may have onto the results of a finite element analysis of a human femur generated from Computer Tomography (CT) data. Starting from a CT dataset of a normal femur, the finite element model was generated using an already validated procedure. The errors associated to the segmentation algorithm adopted, to the CT dataset calibration and to the material assignment procedure were estimated. Simplified loading conditions were applied to the femur to generate a compressive bending and torque loading. The distribution of the mostly used biomechanical indicators was analyzed. The results showed that the level of influence of the input variable is dependent on the load case and on the considered output variable. The coefficients of variation were, however, always smaller than 10%, for all variables.