Evaluation of the feasibility of a novel distance adaptable steady-state visual evoked potential based brain-computer interface

Wu, Chi-Hsu and Lakany, Heba; (2015) Evaluation of the feasibility of a novel distance adaptable steady-state visual evoked potential based brain-computer interface. In: 2015 7th International IEEE/EMBS Conference on Neural Engineering (NER). IEEE, Piscataway, NJ, pp. 57-60. ISBN 9781467363891 (https://doi.org/10.1109/NER.2015.7146559)

[thumbnail of Wu-Lakany-NER-2015-steady-state-visual-evoked-potential-based-brain-computer-interface]
Preview
Text. Filename: Wu_Lakany_NER_2015_steady_state_visual_evoked_potential_based_brain_computer_interface.pdf
Accepted Author Manuscript

Download (371kB)| Preview

Abstract

Steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) has attracted great attention in BCI research due to its advantages over the other electroencephalography (EEG) based BCI paradigms, such as high speed, high signal to noise ratio, high accuracy, commands scalability and minimal user training time. Several studies have demonstrated that SSVEP BCI can provide a reliable channel to the users to communicate and control an external device. While most SSVEP based BCI studies focus on encoding the visual stimuli, enhancing the signal detection and improving the classification accuracy, there is a need to bridge the gap between BCI "bench" research and real world application. This study proposes a novel distance adaptable SSVEP based BCI paradigm which allows its users to operate the system in a range of viewing distances between the user and the visual stimulator. Unlike conventional SSVEP BCI where users can only operate the system at a fixed distance in front of the visual stimulator, users can operate the proposed BCI at a range of viewing distances. 10 healthy subjects participated in the experiment to evaluate the feasibility of the proposed SSVEP BCI. The visual stimulator was presented to the subjects at 4 viewing distances, 60cm, 150cm, 250cm and 350cm. The mean classification accuracy across the subjects and the viewing distances is over 75 The results demonstrate the feasibility of a distance adaptable SSVEP based BCI.