A Tensor-Based Frequency Features Combination Method for Brain–Computer Interfaces

Abstract

With the development of the brain-computer interface (BCI) community, motor imagery-based BCI system using electroencephalogram (EEG) has attracted increasing attention because of its portability and low cost. Concerning the multi-channel EEG, the frequency component is one of the most critical features. However, insufficient extraction hinders the development and application of MI-BCIs. To deeply mine the frequency information, we proposed a method called tensor-based frequency feature combination (TFFC). It combined tensor-to-vector projection (TVP), fast fourier transform (FFT), common spatial pattern (CSP) and feature fusion to construct a new feature set. With two datasets, we used different classifiers to compare TFFC with the state-of-the-art feature extraction methods. The experimental results showed that our proposed TFFC could robustly improve the classification accuracy of about 5% (p < 0.01). Moreover, visualization analysis implied that the TFFC was a generalization of CSP and Filter Bank CSP (FBCSP). Also, a complementarity between weighted narrowband features (wNBFs) and broadband features (BBFs) was observed from the averaged fusion ratio. This article certificates the importance of fre-Manuscript

Enhancing Detection of Control State for High-Speed Asynchronous SSVEP-BCIs Using Frequency-Specific Framework

“…Although the FS framework achieved impressive performance using common machine learning and feature extraction methods, the supervised machine learning that depends on individual training data may result in high costs in practical applications. Future research is needed to combine the FS framework with subject or stimulus transfer methods [45][46][47], a more efficient frequency feature extraction method [48], or the training-free method [49] to reduce the training cost. Towards high-speed and high-accuracy asynchronous SSVEP-BCIs, the performance and data length could be further optimized by incorporating more efficient dynamic stopping strategies [40,41] into the FS framework by determining the data length adaptively for each trial.…”

How Visual Stimuli Evoked P300 is Transforming the Brain–Computer Interface Landscape: A PRISMA Compliant Systematic Review

“…The present study was confined to "Visual Stimuli evoked P300-based BCI". Future work can focus on reviewing other categories of the EEG Signals like SSVEP(s) [148], EEG-EOG [149], miniature-event-related potentials [151], and stimuli like tactical stimuli [150], very small lateral stimuli [151], decoding N2pc components [152], tensor based frequency features combinations [153], etc.…”

Targeting electroencephalography for alcohol dependence: A narrative review

“…The electroencephalogram (EEG) is a popular choice for constructing BCI systems due to its cost‐effectiveness, non‐invasive nature, and portability. Throughout numerous BCI studies, growing attention has been dedicated to analyzing EEG of motor imagery 64–67 . In the field of alcohol dependence, BCIs can be used to monitor the brain activity of patients and detect the craving of alcohol.…”