Investigation of the Effect of Uncertainty in Muscle Pathways on the Knee Joint Contact Force in an Improved Musculoskeletal Model During Squat Motion

Uncertainty about the musculoskeletal modeling results is one of the main issues that are less investigated by users. Although the results of these models are accurate enough for gait activities, high errors were reported for activities with deep knee bending like squatting tasks. Recently, by updating the wrapping surfaces in a powerful musculoskeletal model (Rajagopal), an updated version for deep squatting (Catelli) has been released. But high average errors are reported in simulations using this model for CAMS subjects. Opensim software and Opensim-Matlab API were used. In this study, the effect of uncertainty in the muscle pathways of a scaled version of the Catelli model on the knee contact force was studied using the probabilistic Monte Carlo approach. 43.7±19.02% confidence bounds of KCF error during a squat cycle and 159±77% in deep squat were computed. Therefore, more corrections in muscle pathways parameters in the model specified for simulating squat motion could improve the results.