Optimization of input parameters of an EMG-force model in constant and sinusoidal force contractions

In an electromyographic and muscle force (EMG-force) model, the variability and uncertainty of the input muscle parameters increase the difficulty of assessing this type of model. In this study, a Monte Carlo method is used to evaluate the robustness and the sensitivity of an EMG-force model, recently developed by our team, for two groups of simulations (constant and sinusoidal force contractions). Two existing criteria (EMG/force and force/force-variability relations) and a new criterion derived from this model (root mean square error, Error_RMS, between the force command and the generated force) are used to extract relevant simulations and obtain the optimized parameter ranges in constant force contractions, while only the new criterion could be valuable in sinusoidal force contractions. The comparison of obtained results from the two groups of simulations has shown that the new criterion can replace the two existing criteria in constant and sinusoidal force contractions to give rise to stable optimized input parameter ranges for the studied EMG-force model.