Hybrid Human-Machine Interface to Mouse Control for Severely Disabled People

Abstract

This paper describes a hybrid human-machine interface, based on electro-oculogram (EOG) and electromyogram (EMG), which allows the mouse control of a personal computer using eye movement and the voluntary contraction of any facial muscle. The bioelectrical signals are sensed through adhesives electrodes, and acquired by a custom designed portable and wireless system. The mouse can be moved in any direction, vertical, horizontal and diagonal, by two EOG channels and the EMG signal is used to perform the mouse click action. Blinks are avoided by a decision algorithm and the natural reading of the screen is possible with a specially designed software. A virtual keyboard was used for the experiments with healthy people and with a severely disabled patient. The results demonstrate an intuitive and accessible control, evaluated in terms of performance, time for task execution and user´s acceptance. Besides, a quantitative index to estimate the training impact was computed with good results.