Modelo computacional baseado em interface cérebro-computador para classificação de sinais EEG

Abstract

In contemporary society, the number of people diagnosed with some kind of physical disorder and who present motor impairment can be observed. These disorders affect considerably their life quality and their autonomy. Body control activities performed by human beings utilize neuromuscular traits. People diagnosed with clinical conditions such as Amyotrophic Lateral Sclerosis, Spinal Cord Lesion or Cerebrovascular Accident, for instance, present their neuromuscular traits compromised, and for that reason, they face difficulties regarding mobility. One of the alternatives to overcome that problem is the development of technologies that partially substitute the lost functions of the people with severe motor impairment. In this context the cerebro-computer interfaces, present themselves as a solution to aid individuals with severe motor and neurologic impairment of their daily activities. However, many of these interfaces are constructed isolatedly, thus having their own structures, what generates a negative impact in their reutilization capacity, integration and extension. The general aim of the present investigative paper is to develop a computational model based on a cerebro-computer interface which allows the study and the development of new technologies from electroencephalography signals in order to help individuals with motor impairment in conducting an intelligent wheelchair during their daily activities. Those signals were acquired through a database eegmmidb - EEG Motor Movement/Imagery Dataset. Samples of electroencephalography signals from 106 individuals were utilized in order to validate the computational model. The results obtained in this research work show that the proposed architecture is feasible and reusable for the development of brain-computer interfaces.