Neural network controller for an upper extremity neuroprosthesis

The long term goal of this project is to develop a controller for an upper extremity neuroprosthesis targeted for people with C5/C6 spinal cord injury (SCI). The challenge is to determine how to simultaneously stimulate different paralyzed muscles based on the EMG activity of muscles under retained voluntary control. The proposed controller extracts information from the recorded EMG signals and processes this information to generate the appropriate stimulation levels to activate the paralyzed muscles. The goal of this project was to design and evaluate this controller using a dynamic, three-dimensional musculoskeletal model of the arm. Different arm movements were recorded from able bodied subjects and these kinematics served as input to the model. The model was modified to reflect C5/C6 SCI, and inverse simulations were run to provide muscle activation patterns corresponding to the movements recorded. A set of "voluntary" and "paralyzed" muscles was selected for the controller based on each muscle's relevance as suggested by the simulations. Activation patterns were then used to train a dynamic neural network that predicts "paralyzed" muscle activations from "voluntary" muscle activations. The neural network controller was able to predict the activation level of three paralyzed muscles with less than 2% average prediction error, using four input muscles as inputs