Developing and Testing of an Emg Driven Model to Estimate Ankle Moments and Muscle Forces

Estimation of individual muscle force and its contribution to a joint moment has an important role in understanding injuries and disease. In previous work, we have developed EMG-driven models to estimate isometric forces in muscles about the elbow (Manal et al., 2002). Similarly, Lloyd and Besier (2003) used EMG-driven models to predict dynamic flexion and extension knee moments for a varied range of tasks. This approach has the advantage of estimating muscle force in vivo using EMG signals and an anatomical muscle model as input, and thereby relies on a person’s particular neural control strategy rather than an assumed cost function. Developing the anatomical model is a fundamental step towards good force prediction. The more detailed the model, the more it is able to predict a variety of muscle contractions. Practical limitations, such as the ability to get EMG signals from all muscles, generally leads to a simplification of the model.