Prediction of above-elbow motions in amputees, based on electromyographic(EMG) signals, using nonlinear autoregressive exogenous (NARX) model

Introduction: in order to improve the quality of life of amputees, biomechatronic researchers and biomedical engineers have been trying to use a combination of various techniques to provide suitable rehabilitation systems. diverse biomedical signals, acquired from a specialized organ or cell system, e.g., the nervous system, are the driving force for the whole system. electromyography(emg), as an experimental technique, is concerned with the development, recording, and analysis of myoelectric signals. emg-based research is making progress in the development of simple, robust, user-friendly, and efficient interface devices for the amputees. materials and methods: prediction of muscular activity and motion patterns is a common, practical problem in prosthetic organs. recurrent neural network (rnn) models are not only applicable for the prediction of time series, but are also commonly used for the control of dynamical systems. the prediction can be assimilated to identification of a dynamic process. an architectural approach of rnn with embedded memory is nonlinear autoregressive exogenous (narx) model, which seems to be suitable for dynamic system applications. results: performance of narx model is verified for several chaotic time series, which are applied as input for the neural network. the results showed that narx has the potential to capture the model of nonlinear dynamic systems. the r-value and mse are 8.3×10-3 and 0.99, respectively. conclusion: emg signals of deltoid and pectoralis major muscles are the inputs of the narx network. it is possible to obtain emg signals of muscles in other arm motions to predict the lost functions of the absent arm in aboveelbow amputees, using narx model.