A Novel Digital Speller Based on a Hybrid Brain Computer Interface (hBCI) SSVEP with Eye Tracking

Many individuals with severe muscle atrophy caused by advanced Amyotrophic Lateral Sclerosis (ALS), brain stroke or spinal cord injury have a loss of their communication abilities. For those, who have problems or absence of speech capabilities, a speller system represents more than a tool as sometimes it is the only window that allows interaction with others and with the world. The objective of this work is to evaluate the performance of a speller in Brazilian Portuguese, based on a Novel hybrid Brain Computer Interface (hBCI), which uses Steady State Visual Evoked Potentials (SSVEP) from Electroencephalography (EEG) signals in addition to eye tracking acquired by a camera system with near field infrared light (NIR). The system uses a combination of both informations to ensure that a word is not misspelled. This way, it’s expected a reduction on the time to spell a word and on the user’s frustration, for having to correct mistyped letters. The interface also allows individuals with severe muscle palsy to communicate using only the eyes. This speller is particularly special because until the writing of this article there was no parallel in the Brazilian language. The tests were conducted with three volunteers who were asked to spell four words of everyday use on three trials. The experiment used en Emotiv EPOC + wireless device to obtain EEG signals and a Tobii Eyetracker 4C to obtain eye movements. The results obtained with the system are promising; the participants were able to perform the spelling tasks with an average typing speed of 4.2 characters per second and an accuracy of 80%.