Comparison of knee cartilage biomechanics in finite element analysis using inputs from AnyBody and OpenSim

Abstract Musculoskeletal (MS) modeling is an effective technique for examining body- and joint-level motion and loading. In knee joint models, the obtained joint kinematics and kinetics from the MS analysis are often used as inputs to finite element (FE) analysis to produce estimates for tissue-level responses. Previous research has explored how body- and joint-level estimates might be altered when using different MS modeling software, but there is a lack of information about the possible effects on tissue-level analysis. The aim of this study was to compare the cartilage tissue-level mechanics in subject-specific FE models with kinematic and kinetic inputs obtained from two widely used models, one applied in AnyBody and one in OpenSim. Motion data from gait trials of two healthy subjects were used with two commonly utilized MS models, one from each MS software. The obtained MS estimates were then used as inputs to the FE analysis, which was implemented in ABAQUS. Although the resulting cartilage mechanics were mostly comparable, the AnyBody- driven FE model estimated slightly higher maximum principal stresses in the medial cartilage for both subjects. The OpenSim-driven FE model estimated more anterior and lateral center of pressure in the lateral tibial cartilage, and more posterior in the medial tibial cartilage. Understanding and acknowledging these differences due to changes in the modeling environment is important to ensure reliable and accurate analysis of tissue-level knee mechanics, e.g., when predicting knee osteoarthritis progression and the effects of rehabilitations.