Computational performance analysis of 3D finite element anatomical models using various mesh-to-solid conversions

The finite element method analysis (FEM) is essential for predicting the mechanical behavior of 3D models, but its accuracy depends on the geometric quality of the model. This work proposes a methodology to optimize 3D anatomical models for numerical simulation purposes, applied to a tibia model using four geometric variants by two conversion approaches: faceted and organic geometry. A mesh sensitivity analysis was performed, evaluating computational performance and accuracy using the average von Mises stress. Faceted models performed better than organic models. A model stood out for its accuracy and shorter computing time. It is concluded that the use of faceted geometry and mesh refinement improves the stability and efficiency of the FEM without significantly increasing the computational demand.