TY - JOUR
T1 - Learning Discriminative Spatiospectral Features of ERPs for Accurate Brain-Computer Interfaces
AU - Abibullaev, Berdakh
AU - Zollanvari, Amin
N1 - Funding Information:
Manuscript received June 22, 2018; revised September 27, 2018; accepted November 17, 2018. Date of publication January 16, 2019; date of current version September 4, 2019. This work was partially supported by Nazarbayev University Faculty Development under Grant SOE2018008. (Corresponding author: Berdakh Abibullaev.) B. Abibullaev is with the Department of Robotics and Mechatronics, Nazarbayev University, Astana 010000, Kazakhstan (e-mail:, berdakh.abibullaev@nu.edu.kz).
PY - 2019/9
Y1 - 2019/9
N2 - Constructing accurate predictive models is at the heart of brain-computer interfaces (BCIs) because these models can ultimately translate brain activities into communication and control commands. The majority of the previous work in BCI use spatial, temporal, or spatiotemporal features of event-related potentials (ERPs). In this study, we examined the discriminatory effect of their spatiospectral features to capture the most relevant set of neural activities from electroencephalographic recordings that represent users' mental intent. In this regard, we model ERP waveforms using a sum of sinusoids with unknown amplitudes, frequencies, and phases. The effect of this signal modeling step is to represent high-dimensional ERP waveforms in a substantially lower dimensionality space, which includes their dominant power spectral contents. We found that the most discriminative frequencies for accurate decoding of visual attention modulated ERPs lie in a spectral range less than 6.4 Hz. This was empirically verified by treating dominant frequency contents of ERP waveforms as feature vectors in the state-of-the-art machine learning techniques used herein. The constructed predictive models achieved remarkable performance, which for some subjects was as high as 94% as measured by the area under curve. Using these spectral contents, we further studied the discriminatory effect of each channel and proposed an efficient strategy to choose subject-specific subsets of channels that generally led to classifiers with comparable performance.
AB - Constructing accurate predictive models is at the heart of brain-computer interfaces (BCIs) because these models can ultimately translate brain activities into communication and control commands. The majority of the previous work in BCI use spatial, temporal, or spatiotemporal features of event-related potentials (ERPs). In this study, we examined the discriminatory effect of their spatiospectral features to capture the most relevant set of neural activities from electroencephalographic recordings that represent users' mental intent. In this regard, we model ERP waveforms using a sum of sinusoids with unknown amplitudes, frequencies, and phases. The effect of this signal modeling step is to represent high-dimensional ERP waveforms in a substantially lower dimensionality space, which includes their dominant power spectral contents. We found that the most discriminative frequencies for accurate decoding of visual attention modulated ERPs lie in a spectral range less than 6.4 Hz. This was empirically verified by treating dominant frequency contents of ERP waveforms as feature vectors in the state-of-the-art machine learning techniques used herein. The constructed predictive models achieved remarkable performance, which for some subjects was as high as 94% as measured by the area under curve. Using these spectral contents, we further studied the discriminatory effect of each channel and proposed an efficient strategy to choose subject-specific subsets of channels that generally led to classifiers with comparable performance.
KW - Brain-computer interfaces
KW - EEG
KW - ERPs
KW - P300
KW - machine learning
KW - signal processing
UR - http://www.scopus.com/inward/record.url?scp=85071896214&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85071896214&partnerID=8YFLogxK
U2 - 10.1109/JBHI.2018.2883458
DO - 10.1109/JBHI.2018.2883458
M3 - Article
C2 - 30668507
AN - SCOPUS:85071896214
VL - 23
SP - 2009
EP - 2020
JO - IEEE Journal of Biomedical and Health Informatics
JF - IEEE Journal of Biomedical and Health Informatics
SN - 2168-2194
IS - 5
M1 - 8613780
ER -