Multi‐command SSAEP‐based BCI system with training sessions for SSVEP during an eye fatigue state

Abstract

This paper proposes a steady‐state auditory stimulus modality and a detection algorithm to replace steady‐state visual evoked potential (SSVEP)‐based brain–computer interface (BCI) systems during visual fatigue periods. The optimal speaker position for the steady‐state auditory evoked potential (SSAEP)‐based BCI system and possible electrode positions are investigated. Using the proposed system, an accuracy of 85% for two commands was achieved based on the T3–T5 and T4–T6 electrode positions using only one speaker. SSAEP is a promising BCI modality for mitigating the problem of eye fatigue that often occurs during the use of SSVEP‐based BCI systems. However, SSAEP‐based BCI systems suffer from low accuracy. To increase accuracy, we propose a new enhanced SSAEP training method. The training process was enhanced by instructing users to control their attention levels while simultaneously detecting an auditory stimulus frequency. Furthermore, we propose a corresponding single‐frequency, multi‐command BCI paradigm for the proposed training method. With the proposed paradigm, four commands can be detected using only one auditory stimulus frequency. The proposed training system yielded an accuracy of ∼81% compared to 66% for sessions performed without the proposed training method. © 2017 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Current Status, Challenges, and Possible Solutions of EEG-Based Brain-Computer Interface: A Comprehensive Review

“…The resulting brain response can be localized at the primary auditory cortex (Hill and Schölkopf, 2012). Although the SSAEP-based BCI system yielded promising results, only highly experienced users could maintain the high level of attention needed in order to obtain high accuracy (Punsawad and Wongsawat, 2017).…”

“…However, these are only effective when the frequency bands of the signal do not overlap (Sweeney et al, 2012). In case of spectral overlap, where artifacts are recorded with the EEG, alternative artifact removal techniques are required such as adaptive filtering, Wiener filtering, Bayes filtering (Sweeney et al, 2012), surface Laplacian transforms (Fitzgibbon et al, 2013), regression (Gratton et al, 1983), Common Average Referencing (CAR) (Zaizu Ilyas et al, 2015), EOG correction (Croft and Barry, 2000), and blind source separation (BSS) (Oosugi et al, 2017), as well as more modern attempts, for instance, the wavelet transform (WT) method (Punsawad and Wongsawat, 2017), empirical mode decomposition (EMD) (Zhang et al, 2008), Canonical Correlation Analysis (CCA) (de Clercq et al, 2006), and nonlinear mode decomposition (NMD) (Iatsenko et al, 2015). It is worth noting that the BSS methods are also called componentbased techniques, as they employ principal component analysis (PCA) or independent component analysis (ICA).…”

Quantitative Formula of Blink Rates-Pupillometry for Attention Level Detection in Supervised Machine Learning

“…By extracting quantitative EEG (QEEG) features from an EEG signal, as well as heart rate and heart rate variability (HRV) from an ECG, that researcher found evidence of differing activity in the engagement and disengagement states, in both the EEG and ECG. Other attention level detection has also done by using SSVEP [6], the attention is categorized based on EEG signal, when alpha ratio is decreased and beta ratio is increased than baseline. That study got classification accuracy based on algorithm is 81 %.…”

Supervising SSSEP Experiments with a Bluetooth Android Remote Control Application

“…Electrodes are positioned mainly on the temporal part of the human scalp in order to detect SSAEP. Steady state auditory evoked potential can be used to replace steady state visual evoked potential in brain-computer interface systems during visual fatigue periods [13]. Auditory ERP can also be used in auditory speller BCI or multichoice based BCI [14] [15].…”