A step towards home-based robotic rehabilitation: An interface circuit for EEG/SEMG actuated orthosis

The effectiveness of rehabilitation will be increased substantially (e.g. stroke patients) if the patients are able to use a robotic rehabilitation system at home, after having trained on it at the hospital. Due to high cost and complex architecture, most robotic orthoses are limited to use in the hospital. The "active" orthoses that make use of bio-signals for control purposes, are at present limited in their versatility, portability and usability. At the same time, studies show that rehabilitation speeds up when the level of patient engagement is higher. To make home-use a reality, it is of paramount importance that the system is low-cost, portable, reliable and simple to operate. An acquisition and control system which satisfies these goals will create a significant impact on patient adoption of robotic rehabilitation devices. The sub-system design that is described in this paper is part of a wider research work to develop an accelerated stroke rehabilitation platform utilizing an EEG/SEMG based upper extremity robotic orthosis. This sub-system forms the \´interface\´ between the patient and the computer / controlling device used for signal processing and orthosis control. Cost and weight is reduced significantly. The circuit can interface with industry standard data acquisition devices and switch seamlessly between surface electromyography (SEMG) and electroencephelography (EEG) operation. Test results are presented both with simulated signals as well as actual SEMG signals.