Physiologically based boundary conditions in finite element modelling
Finite element analysis has been used extensively in the study of bone loading and implant performance, such as in the femur. The boundary conditions applied vary widely, generally producing excessive femoral deformation, and although it has been shown that the muscle forces influence femoral deflections and loading, little consideration has been given to the displacement constraints. It is hypothesised that careful application of physiologically based constraints can produce physiological deformation, and therefore straining, of the femur. Joint contact forces and a complete set of muscle forces were calculated based on the geometry of the Standardised Femur using previously validated musculoskeletal models. Five boundary condition cases were applied to a finite element model of the Standardised Femur: (A) diaphyseally constrained with hip contact and abductor forces; (B) case A plus vasti forces; (C) case A with complete set of muscle forces; (D) distally constrained with all muscle forces; (E) physiological constraints with all muscle forces. It was seen that only the physiological boundary conditions, case E, produced physiological deflections (<2.0 mm) of the femoral head in both the coronal and sagittal planes, which resulted in minimal reaction forces at the constrained nodes. Strains in the mid-diaphysis varied by up to 600 μ-strain under walking loads and 1000 μ-strain under stair climbing loads. The mode of loading, as indicated by the strain profiles on the cortex also varied substantially under these boundary conditions, which has important consequences for studies that examine localised bone loading such as fracture or bone remodelling simulations.
|Keywords||Constraints, Femoral loading, Finite element, Physiological boundary conditions|
|Journal||Journal of Biomechanics|
Speirs, A, Heller, M.O. (Markus O.), Duda, G.N. (Georg N.), & Taylor, W.R. (William R.). (2007). Physiologically based boundary conditions in finite element modelling. Journal of Biomechanics, 40(10), 2318–2323. doi:10.1016/j.jbiomech.2006.10.038