Enhancing Detection of SSVEPs for a High-Speed Brain Speller Using Task-Related Component Analysis

Abstract

Objective: This study proposes and evaluates a novel data-driven spatial filtering approach for enhancing steady-state visual evoked potentials (SSVEPs) detection toward a high-speed brain-computer interface (BCI) speller.

Methods: Task-related component analysis (TRCA), which can enhance reproducibility of SSVEPs across multiple trials, was employed to improve the signal-to-noise ratio (SNR) of SSVEP signals by removing background electroencephalographic (EEG) activities. An ensemble method was further developed to integrate TRCA filters corresponding to multiple stimulation frequencies. This study conducted a comparison of BCI performance between the proposed TRCA-based method and an extended canonical correlation analysis (CCA)-based method using a 40-class SSVEP dataset recorded from 12 subjects. An online BCI speller was further implemented using a cue-guided target selection task with 20 subjects and a free-spelling task with 10 of the subjects.

Results: The offline comparison results indicate that the proposed TRCA-based approach can significantly improve the classification accuracy compared with the extended CCA-based method. Furthermore, the online BCI speller achieved averaged information transfer rates (ITRs) of 325.33 ± 38.17 bits/min with the cue-guided task and 198.67 ± 50.48 bits/min with the free-spelling task.

Conclusion: This study validated the efficiency of the proposed TRCA-based method in implementing a high-speed SSVEP-based BCI.

Significance: The high-speed SSVEP-based BCIs using the TRCA method have great potential for various applications in communication and control.

Fig. 1

Stimulus design of the 40-target BCI system. Frequency and phase values specified for each target.

Fig. 2

Diagram of the task-related component analysis (TRCA) in SSVEP analysis.

Fig. 3

Diagrams of the proposed methods. (a) The TRCA-based method and (b) The ensemble TRCA-based method. 1-D and 2-D correlation analyses were used in (a) and (b), respectively.

Fig. 4

Performance comparison of three methods using recorded SSVEPs. (a) Averaged classification accuracy, (b) simulated ITRs across subjects as a function of data length (d). (c) An example of r-square values for SSVEPs at 9.6 Hz for each method (data length 300 ms). The error bars indicate standard errors. The asterisks in the subfigure (a) and (b) indicate significant difference between the three methods obtained by one-way repeated measures ANOVAs, and those in the subfigure (c) indicate significant difference between each pair of the two methods by paired t-tests (*p<0.05, **p<0.01, ***p<0.001).

Fig. 5

Averaged classification accuracy with (a) different numbers of training trials (N_t), (b) different numbers of electrodes (N_c), and (c) different numbers of sub-bands (N_m) across subjects with 300ms-long epochs. The error bars indicate standard errors.