Filter Bank-Driven Multivariate Synchronization Index for Training-Free SSVEP BCI

Abstract

In recent years, multivariate synchronization index (MSI) algorithm, as a novel frequency detection method, has attracted increasing attentions in the study of brain-computer interfaces (BCIs) based on steady state visual evoked potential (SSVEP). However, MSI algorithm is hard to fully exploit SSVEPrelated harmonic components in the electroencephalogram (EEG), which limits the application of MSI algorithm in BCI systems. In this paper, we propose a novel filter bank-driven MSI algorithm (FBMSI) to overcome the limitation and further improve the accuracy of SSVEP recognition. We evaluate the efficacy of the FBMSI method by developing a 6-command SSVEP-NAO robot system with extensive experimental analyses. An offline experimental study is first performed with EEG collected from nine subjects to investigate the effects of varying parameters on the model performance. Offline results show that the proposed method has achieved a stable improvement effect. We further conduct an online experiment with six subjects to assess the efficacy of the developed FBMSI algorithm in a real-time BCI application. The online experimental results show that the FBMSI algorithm yields a promising average accuracy of 83.56% using a data length of even only one second, which was 12.26% higher than the standard MSI algorithm. These extensive experimental results confirmed the effectiveness of the FBMSI algorithm in SSVEP recognition and demonstrated its potential application in the development of improved BCI systems.

Enhancing Performance of SSVEP-Based Visual Acuity via Spatial Filtering

“…Because of the fast and accurate requirement and infrequent testing for visual acuity assessment (Zheng et al, 2021), the trainingfree methods were adequate here. The filter bank strategy in training-free methods, such as filter bank CCA (FBCCA) (Chen et al, 2015) and filter bank MSI (FBMSI) (Qin et al, 2021), may be also used to enhance the performance of SSVEP-based visual acuity assessment in future work. In contrast, the subjectspecific training methods with the best performance (Zerafa et al, 2018), requiring training data from the specific user and needing the cost of long and tiring training sessions, such as individual template-based CCA (itCCA) (Bin et al, 2011), combined-CCA (Nakanishi et al, 2014;Wang et al, 2014b), multiway CCA (Zhang et al, 2011), multiset CCA (Zhang et al, 2014b), and task-related component analysis (TRCA) (Nakanishi et al, 2018a), may be more suitable for the situation where the subjects need long-term use of BCI system, such as the vision training with SSVEP biofeedback in amblyopia (Lapajne et al, 2020).…”

An Electric Wheelchair Manipulating System Using SSVEP-Based BCI System

“…MSI is an algorithm that can be directly analyzed without the training that CCA needs. This measure is to estimate the synchronization between the actual mixed signals and the reference signals as a potential index for recognizing the stimulus frequency [ 17 , 27 ]. We use the same filtering method as CCA to process the EEG signals before using MSI for analysis.…”

A CNN-Based Deep Learning Approach for SSVEP Detection Targeting Binaural Ear-EEG

“…The most widely used methods for these enhanced CCA include the following: combination method-CCA (Nakanishi et al, 2015 ), individual template CCA (IT-CCA) (Wang et al, 2014 ), and more recently proposed task-related components analysis (TRCA) (Nakanishi et al, 2018 ). Most recent work on SSVEP includes advanced techniques such as filter bank-driven multivariate synchronization algorithm (Qin et al, 2021 ) and multivariate variational mode decomposition-informed canonical correlation analysis (Chang et al, 2022 ).…”