A one-parameter neural activation to muscle activation model: estimating isometric joint moments from electromyogram

Nonlinearities have been observed in the isometric EMG–force relationship. However, these are generally not included when using EMG-driven Hill-type muscle models that account for muscle activation dynamics. In this paper, we present a formulation for a one-parameter transformation model (i.e., A-model) that accounts for the type of physiological nonlinearities observed at low levels of force. The general shape for the curvilinear portion of the curve was based on phenomenological data reported by Woods and Bigland-Ritchie. The one-parameter A-model is easy to implement, and when used with an EMG-driven Hill-type model, was shown to provide a better fit of the measured joint moment. Optimization methods were used to determine the appropriate curvature of the relationship for each muscle, and thus introduced a degree of “tuning” to each subject.