Kosmyna, N., Balyan, A., Hauptmann, E. (2023). Decoding Visual Imagery Using EEG/EOG Glasses: A Pilot Study. In: Arai, K. (eds) Proceedings of the Future Technologies Conference (FTC) 2022, Volume 2. FTC 2022 2022. Lecture Notes in Networks and Systems, vol 560. Springer, Cham. https://doi.org/10.1007/978-3-031-18458-1_29

Abstract

 Various studies have explored new and intuitive paradigms for Brain Computer Interfaces (BCIs) as control strategies such as visual imagery. Recent literature has also investigated the use of socially acceptable, mobile and miniature devices for BCIs. This study explores the possibility of using a mobile EEG/EOG device in a form-factor of the glasses for decoding visual imagery that could be used as a control strategy for BCIs. The decoding performance of a mobile EEG/EOG device was compared to a 16 channel research-grade EEG headset. 14 participants were asked to mentally visualize the image they were presented belonging to one of two different groups: coffee brands and artworks. Two devices simultaneously recorded EEG data from both mobile and consumer grade EEG headset as well as EOG data from the glasses during the experiment. Common spatial pattern and sample entropy were used as a feature set and a Linear Discriminant Analysis (LDA) classifier was used for the classification of tasks. The low channel EEG/EOG device achieved an average accuracy of 73% and 75% for coffee and art stimuli groups, respectively. The research grade EEG system achieved an average accuracy of 52% and 79% for the same families of stimuli. The results indicate that low channel EEG/EOG devices could be further explored for decoding visual imagery.