Surface electromyography recording of spontaneous eyeblinks : applications in neuroprosthetics

Objective. We are designing an implantable neuroprosthesis for the treatment of unilateral facial paralysis. The envisioned biomimetic device paces artificial blinks in the paretic eyelid when activity in the healthy orbicularis oculi (orbicularis) muscle is detected. The present article focuses on electromyography (EMG)–based eyeblink detection. Study Design. A pilot clinical study was performed in healthy volunteers who were intended to represent individuals with facial paralysis. Spontaneous eyeblinks were detected by a surface EMG recording. Blink detection accuracy was tested at rest and during voluntary smiling and chewing. Setting. Fifteen participants were asked to wear surface recording electrodes on the left side of their face, detecting the orbicularis oculi, the masseter, and the zygomatic muscle EMG activity. Subjects and Methods. Participants were asked to look ahead, voluntarily smile, and chew according to an experimental protocol. Custom software was designed with the purpose of selectively filtering the multichannel EMG recordings and triggering a digital output. Results. The software filter allowed elimination of spurious artificial eyeblinks and thus increased the accuracy of the EMG recording apparatus for the spontaneous blinking. Conclusion. Orbicularis oculi EMG recording worked as a realtime eyeblink-detecting system. Moreover, the multichannel EMG recording coupled to a proper digital signal processing was very effective in specifically detecting the spontaneous blinking during other facial muscle activities. With regard to closed loop biomimetic devices for the pacing of the eyeblink, the EMG signal represents a valid option for the recording side.