Device control using gestures sensed from EMG

In this paper, we present neuro-electric interfaces for virtual device control. The examples presented rely upon sampling electromyogram data from a participant's forearm. This data is then set into pattern recognition software that had been trained to distinguish gestures from a given gesture set. The pattern recognition software consists of hidden Markov models, which are used to recognize the gestures as they are being performed in real-time. two experiments were conducted to examine the feasibility of this interface technology. The first replicate a virtual joystick interface and the second replicated a keyboard.