State estimation of walking phase and functional electrical stimulation by wearable device

Functional electrical stimulation (FES) is useful to improve the gait of patients with peroneal nerve palsy or spastic hemiparesis after stroke. So as to apply FES to such patients, we have to have estimators for detecting the timing of phase switching in walking motion. We designed a wearable device for state estimating of walking and functional electrical stimulation. We consider the implementation of artificial neural network (ANN) into the device, and propose a method for supervised learning of the ANN. Two experiments have been conducted to show the effectiveness of the wearable device. The accuracy of estimating the timing for FES is good enough for the practical application.