Cerebral palsy-affected individuals' brain-computer interface for wheelchair movement in an indoor environment using mental tasks

Abstract

The technique of measuring brain signals or activities by placing electrodes on the scalp of human beings is called Electroencephalogram (EEG). Brain-computer interface (BCI) is a technique to capture brain signals and translate them into control signals to run external devices. With the combination of these two techniques, we can create BCI using brain signals. Methods: In this study, the author considered conducting two types of methods offline and online both in the indoor environment using the Fast Fourier Transform (FFT) technique with Feed Forward Neural Network trained with Bat optimization algorithm (FFNNBOA). The study was carried out on two different age groups between 30 to 45 years and 46 to 60 years with four different tasks. Based on the execution of the four different tasks concerning two different age groups, the accuracy obtained during classification is 94.35 % and 93.76 % for offline and online modes. Results: The results it is observed that the classification accuracy for the age group belonging 46 to 60 is comparably higher than that of the conventional classification model. The offline and online tests were conducted for both age groups persons and obtained the recognizing accuracy of 95 %, 93.25 %, and 93.75 %, 91.75 % for the two modes. This study confirms that the performances of the subjects belonging to age groups 30 to 45 are higher than the age groups belonging to 46 to 60 in terms of classification, offline, and online mode. Finally, this study also identified that subject S4 from the 30 to 45 age group showed 100 % accuracy in both offline and online signal acquisition.