Enhancing Detection of SSMVEP Induced by Action Observation Stimuli Based on Task-Related Component Analysis

Abstract

Action observation (AO)-based brain-computer interface (BCI) is an important technology in stroke rehabilitation training. It has the advantage of simultaneously inducing steady-state motion visual evoked potential (SSMVEP) and activating sensorimotor rhythm. Moreover, SSMVEP could be utilized to perform classification. However, SSMVEP is composed of complex modulation frequencies. Traditional canonical correlation analysis (CCA) suffers from poor recognition performance in identifying those modulation frequencies at short stimulus duration. To address this issue, task-related component analysis (TRCA) was utilized to deal with SSMVEP for the first time. An interesting phenomenon was found: different modulated frequencies in SSMVEP distributed in different task-related components. On this basis, a multi-component TRCA method was proposed. All the significant task-related components were utilized to construct multiple spatial filters to enhance the detection of SSMVEP. Further, a combination of TRCA and CCA was proposed to utilize both advantages. Results showed that the accuracies using the proposed methods were significant higher than that using CCA at all window lengths and significantly higher than that using ensemble-TRCA at short window lengths (≤2 s). Therefore, the proposed methods further validate the induced modulation frequencies and will speed up the application of the AO-based BCI in rehabilitation.

An extended binary subband canonical correlation analysis detection algorithm oriented to the radial contraction-expansion motion steady-state visual evoked paradigm

“…Majority of existing research on the radial contraction-expansion motion paradigm includes comparative studies of EEG characteristics evoked by different paradigms and practical research on the radial contraction-expansion motion paradigm. The identification algorithms used are mostly canonical correlation analysis (CCA) [8]. As a statistical method for measuring the potential correlation between two multidimensional variables, CCA is widely used in SSVEP-BCI [9].…”

DF-SSmVEP: Dual Frequency Aggregated Steady-State Motion Visual Evoked Potential Design with Bifold Canonical Correlation Analysis

“…Literature Review: Generally speaking, there are two main visual BCI Paradigms, (1) Steady-State Visually Evoked Potential (SSVEP) [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ], where light flashing (flickering) visual stimulus is used to induce evoked potentials in the EEG signals, and; (2) Steady-State motion Visual Evoked Potentials (SSmVEP) [ 15 , 16 , 17 , 18 , 19 ], where instead of using flickering, some form of graphical motion is used to evoke potentials. The former category (SSVEP) has been the main research theme due to its high achievable Information Transfer Rate (ITR), minimal requirement for user training, and excellent interactive potentials, such as high tolerance to artifacts and robust performance across users.…”