Steady-state Visual Evoked Potential Data Research Articles

SSVEP-DAN: Cross-Domain Data Alignment for SSVEP-Based Brain-Computer Interfaces.

Steady-state visual-evoked potential (SSVEP)-based brain-computer interfaces (BCIs) offer a non-invasive means of communication through high-speed speller systems. However, their efficiency is highly dependent on individual training data acquired during time-consuming calibration sessions. To address the challenge of data insufficiency in SSVEP-based BCIs, we introduce SSVEP-DAN, the first dedicated neural network model designed to align SSVEP data across different domains, encompassing various sessions, subjects, or devices. Our experimental results demonstrate the ability of SSVEP-DAN to transform existing source SSVEP data into supplementary calibration data. This results in a significant improvement in SSVEP decoding accuracy while reducing the calibration time. We envision SSVEP-DAN playing a crucial role in future applications of high-performance SSVEP-based BCIs. The source code for this work is available at: https://github.com/CECNL/SSVEP-DAN.

Using Determinant Point Process in Generative Adversarial Networks for SSVEP Signals Synthesis.

Steady-state visual evoked potential (SSVEP) is one of the main paradigms of brain-computer interface (BCI). However, the acquisition method of SSVEP can cause subject fatigue and discomfort, leading to the insufficiency of SSVEP databases. Inspired by generative determinantal point process (GDPP), we utilize the determinantal point process in generative adversarial network (GAN) to generate SSVEP signals. We investigate the ability of the method to synthesize signals from the Benchmark dataset. We further use some evaluation metrics to verify its validity. Results prove that the usage of this method significantly improved the authenticity of generated data and the accuracy (97.636%) of classification using deep learning in SSVEP data augmentation.

A transformer-based deep neural network model for SSVEP classification

Steady-state visual evoked potential (SSVEP) is one of the most commonly used control signals in the brain–computer interface (BCI) systems. However, the conventional spatial filtering methods for SSVEP classification highly depend on the subject-specific calibration data. The need for the methods that can alleviate the demand for the calibration data becomes urgent. In recent years, developing the methods that can work in inter-subject scenario has become a promising new direction. As a popular deep learning model nowadays, Transformer has been used in EEG signal classification tasks owing to its excellent performance. Therefore, in this study, we proposed a deep learning model for SSVEP classification based on Transformer architecture in inter-subject scenario, termed as SSVEPformer, which was the first application of Transformer on the SSVEP classification. Inspired by previous studies, we adopted the complex spectrum features of SSVEP data as the model input, which could enable the model to simultaneously explore the spectral and spatial information for classification. Furthermore, to fully utilize the harmonic information, an extended SSVEPformer based on the filter bank technology (FB-SSVEPformer) was proposed to improve the classification performance. Experiments were conducted using two open datasets (Dataset 1: 10 subjects, 12 targets; Dataset 2: 35 subjects, 40 targets). The experimental results show that the proposed models could achieve better results in terms of classification accuracy and information transfer rate than other baseline methods. The proposed models validate the feasibility of deep learning models based on Transformer architecture for SSVEP data classification, and could serve as potential models to alleviate the calibration procedure in the practical application of SSVEP-based BCI systems.

A survey of deep learning-based classification methods for steady-state visual evoked potentials

Purpose Steady-state visual evoked potential (SSVEP) based BCI has attracted great interests owing to the high information transfer rate (ITR) and little training requirement. The performance of SSVEP-based BCI heavily depends on the classification methods. Deep Learning (DL) technology provides an alternative avenue for the data classification in SSVEP-based BCI, and has received increasing interests in recent years. This review aimed to summarize the progress of DL-based classification methods for SSVEP data over the past decade.Materials and method The literature was searched and selected based on the research topics of DL and SSVEP. We categorized these methods into four classes, i.e., traditional neural network structures-based DL methods, traditional frequency recognition methods inspiring DL methods, attention mechanisms-based DL models, and transfer learning technology-based DL methods, and generative model-based recognition method. Moreover, we analyzed the current challenges and presented future research opportunities.Conclusions This study provides a systematic description of the current development status on DL-based SSVEP classification methods, and sheds insight on future researches.

Cross-Subject Transfer Method Based on Domain Generalization for Facilitating Calibration of SSVEP-Based BCIs.

In steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs), various spatial filtering methods based on individual calibration data have been proposed to alleviate the interference of spontaneous activities in SSVEP signals for enhancing the SSVEP detection performance. However, the time-consuming calibration session would increase the visual fatigue of subjects and reduce the usability of the BCI system. The key idea of this study is to propose a cross-subject transfer method based on domain generalization, which transfers the domain-invariant spatial filters and templates learned from source subjects to the target subject with no access to the EEG data from the target subject. The transferred spatial filters and templates are obtained by maximizing the intra- and inter-subject correlations using the SSVEP data corresponding to the target and its neighboring stimuli. For SSVEP detection of the target subject, four types of correlation coefficients are calculated to construct the feature vector. Experimental results estimated with three SSVEP datasets show that the proposed cross-subject transfer method improves the SSVEP detection performance compared to state-of-art methods. The satisfactory results demonstrate that the proposed method provides an effective transfer learning strategy requiring no tedious data collection process for new users, holding the potential of promoting practical applications of SSVEP-based BCI.

Improving Generalized Zero-Shot Learning SSVEP Classification Performance From Data-Efficient Perspective.

Generalized zero-shot learning (GZSL) has significantly reduced the training requirements for steady-state visual evoked potential (SSVEP) based brain-computer interfaces (BCIs). Traditional methods require complete class data sets for training, but GZSL allows for only partial class data sets, dividing them into 'seen' (those with training data) and 'unseen' classes (those without training data). However, inefficient utilization of SSVEP data limits the accuracy and information transfer rate (ITR) of existing GZSL methods. To this end, we proposed a framework for more effective utilization of SSVEP data at three systematically combined levels: data acquisition, feature extraction, and decision-making. First, prevalent SSVEP-based BCIs overlook the inter-subject variance in visual latency and employ fixed sampling starting time (SST). We introduced a dynamic sampling starting time (DSST) strategy at the data acquisition level. This strategy uses the classification results on the validation set to find the optimal sampling starting time (OSST) for each subject. In addition, we developed a Transformer structure to capture the global information of input data for compensating the small receptive field of existing networks. The global receptive fields of the Transformer can adequately process the information from longer input sequences. For the decision-making level, we designed a classifier selection strategy that can automatically select the optimal classifier for the seen and unseen classes, respectively. We also proposed a training procedure to make the above solutions in conjunction with each other. Our method was validated on three public datasets and outperformed the state-of-the-art (SOTA) methods. Crucially, we also outperformed the representative methods that require training data for all classes.

Improving the Classification of Real-World SSVEP Data in Brain-Computer Interface Speller Systems Using Deep Convolutional Neural Networks

Purpose: Brain-Computer Interface (BCI) Speller systems help people with mobility impairments improve their cognitive and physical abilities. Steady-State Visual Evoked Potential (SSVEP) signals have been used to build high-speed BCI speller systems. SSVEP signals are a subtype of Visual Evoked Potential (VEP), a form of co-frequency, and the harmonics response elicited by a specific frequency stimulus. Noise and artifacts are critical issues for target detection in SSVEP-based BCI systems.
 Materials and Methods: Thus, it is essential to provide target detection techniques that operate well in the presence of noises. One solution for overcoming the noise impact is to employ approaches that automatically extract the appropriate features for target detection from the training data. Deep Convolutional Neural Network (DCNN) was utilized in this study to automatically extract features from SSVEP data in noisy conditions. Moreover, the BETA database, which contains SSVEP data from 70 individuals collected outside of the electromagnetic shielding room, was used. In this regard, a suitable DCNN structure for target stimulus frequency identification was first designed. The network was pre-trained with part of the data from the BETA database. Finally, at the single-subject level, this pre-trained network was retrained and evaluated.
 Results: The results showed that after retraining, the accuracy and Information Transfer Rate (ITR) increased (p-value < 0.01) for all participants.
 Conclusion:.The enhancement in accuracy and ITR are 25.72% and 43.10 bpm, respectively.

Incorporating Neighboring Stimuli Data for Enhanced SSVEP-Based BCIs

Various spatial filters have been proposed to enhance the target identification performance of steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs). The current methods only extract the target-related information from the corresponding stimulus to learn the spatial filter parameter. However, the SSVEP data from neighboring stimuli also contain frequency information of the target stimulus, which could be utilized to further improve the target identification performance. In this paper, we propose a new method incorporating SSVEPs from the neighboring stimuli to strengthen the target-related frequency information. First, The spatial filter is obtained by maximizing the summation of covariances of SSVEP data corresponding to the target and its neighboring stimuli. Then the correlation features between spatially filtered templates and test data are calculated for target detection. For the performance evaluation, we implemented the offline experiment using the 40-class benchmark dataset from 35 subjects and the 12-target self-collected dataset from 11 subjects. Compared with the state-of-art spatial filtering methods, the proposed method showed superiority in classification accuracy and information transfer rate (ITR). The comparison results demonstrate the effectiveness of the proposed spatial filter for target identification in SSVEP-based BCIs.

KDLPCCA-Based Projection for Feature Extraction in SSVEP-Based Brain-Computer Interfaces

An electroencephalogram (EEG) signal projection using kernel discriminative locality preserving canonical correlation analysis (KDLPCCA)-based correlation with steady-state visual evoked potential (SSVEP) templates for frequency recognition is presented in this paper. With KDLPCCA, not only a non-linear correlation but also local properties and discriminative information of each class sample are considered to extract temporal and frequency features of SSVEP signals. The new projected EEG features are classified with classical machine learning algorithms, namely, K-nearest neighbors (KNNs), naive Bayes, and random forest classifiers. To demonstrate the effectiveness of the proposed method, 16-channel SSVEP data corresponding to 4 frequencies collected from 5 subjects were used to evaluate the performance. Compared with the state of the art canonical correlation analysis (CCA), experimental results show significant improvements in classification accuracy and information transfer rate (ITR), achieving 100% and 240 bits/min with 0.5 s sample block. The superior performance demonstrates that this method holds the promising potential to achieve satisfactory performance for high-accuracy SSVEP-based brain-computer interfaces.

Optimal Approach for Signal Detection in Steady-State Visual Evoked Potentials in Humans Using Single-Channel EEG and Stereoscopic Stimuli.

In EEG studies, one of the most common ways to detect a weak periodic signal in the steady-state visual evoked potential (SSVEP) is spectral evaluation, a process that detects peaks of power present at notable temporal frequencies. However, the presence of noise decreases the signal-to-noise ratio (SNR), which in turn lowers the probability of successful detection of these spectral peaks. In this paper, using a single EEG channel, we compare the detection performance of four different metrics to analyse the SSVEP: two metrics that use spectral power density, and two other metrics that use phase coherency. We employ these metrics find weak signals with a known temporal frequency hidden in the SSVEP, using both simulation and real data from a stereoscopic apparent depth movement perception task. We demonstrate that out of these metrics, the phase coherency analysis is the most sensitive way to find weak signals in the SSVEP, provided that the phase information of the stimulus eliciting the SSVEP is preserved.

An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces.

Brain-computer interfaces (BCIs) provide humans a new communication channel by encoding and decoding brain activities. Steady-state visual evoked potential (SSVEP)-based BCI stands out among many BCI paradigms because of its non-invasiveness, little user training, and high information transfer rate (ITR). However, the use of conductive gel and bulky hardware in the traditional Electroencephalogram (EEG) method hinder the application of SSVEP-based BCIs. Besides, continuous visual stimulation in long time use will lead to visual fatigue and pose a new challenge to the practical application. This study provides an open dataset, which is collected based on a wearable SSVEP-based BCI system, and comprehensively compares the SSVEP data obtained by wet and dry electrodes. The dataset consists of 8-channel EEG data from 102 healthy subjects performing a 12-target SSVEP-based BCI task. For each subject, 10 consecutive blocks were recorded using wet and dry electrodes, respectively. The dataset can be used to investigate the performance of wet and dry electrodes in SSVEP-based BCIs. Besides, the dataset provides sufficient data for developing new target identification algorithms to improve the performance of wearable SSVEP-based BCIs.

Muscle Artifact Removal Toward Mobile SSVEP-Based BCI: A Comparative Study

Steady-state visual evoked potential (SSVEP) serves as one of the extensively utilized paradigms for brain–computer interface (BCI). SSVEP-based BCI has advantages of good classification accuracy, high information transfer rate (ITR), and little user training. However, there exist few studies investigating SSVEP-based BCI in the mobile situation, where electroencephalography (EEG) data are inevitably contaminated by muscle activities, and meanwhile, a limited number of channels are preferred. In this study, two major muscle artifact removal schemes are investigated in the dynamic scenario. One scheme adopts state-of-the-art blind source separation (BSS) methods without electromyogram (EMG) reference, while the other utilizes auxiliary EMG channels as a reference in the denoising process. To evaluate the performance of two schemes, we have tested them on real EEG data collected from ten subjects. The results demonstrate that, in the few-channel condition of SSVEP data, there are classification accuracy improvements of 9.93% for the canonical correlation analysis (CCA) indicator and 15.72% for the task-related component analysis (TRCA) indicator after processed by recursive least-squares (RLS) filtering with auxiliary EMG, and in the single-channel condition, there are classification accuracy improvements of 22.00% for the CCA indicator and 24.42% for the TRCA indicator after RLS filtering, averaged across all the subjects and data lengths. However, the BSS-based methods fail to improve the classification accuracies of SSVEP. Thus, it is of necessity to mount additional EMG channels for effective muscle activity suppression. This study provides an insightful suggestion for future real-life BCI applications.

EEGNet With Ensemble Learning to Improve the Cross-Session Classification of SSVEP Based BCI From Ear-EEG

Ear-electroencephalography (ear-EEG) using electrodes placed above hairless areas around ears is a convenient and comfortable method for signal recording in practical applications of steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI). However, due to the constraint of electrode distribution behind the ear, the amplitude of SSVEP in ear-EEG signals is relatively low, which hinders the application of ear-EEG in SSVEP-based BCI. This study was aimed to improve the performance of ear-EEG in SSVEP-based BCI through re-implementing a compact convolutional neural network (EEGNet) with ensemble learning. We first evaluated the feasibility of applying widely used EEGNet models with different kernel numbers to decode SSVEP in ear-EEG signals. Then we applied an ensemble learning strategy to combine EEGNet models with different kernel numbers to improve the classification of ear-EEG signals. The ear-EEG data was from an open dataset, which acquired three sessions of SSVEP data induced by three flicker stimuli from eleven subjects. The average accuracy of EEGNet with ensemble learning for ear-EEG signals in cross-session validations at 1 s window length was 81.12% (from session 1 to session 2) and 81.74% (from session 1 to session 3), which significantly outperformed canonical correlation analysis (CCA). In addition, the network visualization indicated that EEGNet extracted features related to stimulation frequencies. The results showed promise for accurate classification of SSVEP in ear-EEG signals using deep learning models with strategies, helping to promote the SSVEP based BCI from laboratory to practical application.

In-Ear Electrode EEG for Practical SSVEP BCI

Steady State Visual Evoked Potential (SSVEP) methods for brain–computer interfaces (BCI) are popular due to higher information transfer rate and easier setup with minimal training, compared to alternative methods. With precisely generated visual stimulus frequency, it is possible to translate brain signals into external actions or signals. Traditionally, SSVEP data is collected from the occipital region using electrodes with or without gel, normally mounted on a head cap. In this experimental study, we develop an in-ear electrode to collect SSVEP data for four different flicker frequencies and compare against occipital scalp electrode data. Data from five participants demonstrates the feasibility of in-ear electrode based SSVEP, significantly enhancing the practicability of wearable BCI applications.

A CNN-based compare network for classification of SSVEPs in human walking.

Brain-computer interface (BCI) can provide a way for the disabled to interact with the outside world. Steady-state visual evoked potential (SSVEP), which evokes potential through visual stimulation is one of important BCI paradigms. In laboratory environment, the classification accuracy of SSVEPs is excellent. However, in motion state, the accuracy will be greatly affected and reduce quite a lot. In this paper, in order to improve the classification accuracy of the SSVEP signals in the motion state, we collected SSVEP data of five targets at three speeds of 0km/h, 2.5km/h and 5km/h. A compare network based on convolutional neural network (CNN) was proposed to learn the relationship between EEG signal and the template corresponding to each stimulus frequency and classify. Compared with traditional methods (i.e., CCA, FBCCA and SVM) and state-of-the-art method (CNN) on the collected SSVEP datasets of 20 subjects, the method we proposed always performed best at different speeds. Therefore, these results validated the effectiveness of the method. In addition, compared with the speed of 0 km / h, the accuracy of the compare network at a high walking rate (5km/h) did not decrease much, and it could still maintain a good performance.

A Training Data-Driven Canonical Correlation Analysis Algorithm for Designing Spatial Filters to Enhance Performance of SSVEP-Based BCIs.

Canonical correlation analysis (CCA) is an effective spatial filtering algorithm widely used in steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs). In existing CCA methods, training data are used for constructing templates of stimulus targets and the spatial filters are created between the template signals and a single-trial testing signal. The fact that spatial filters rely on testing data, however, results in low classification performance of CCA compared to other state-of-the-art algorithms such as task-related component analysis (TRCA). In this study, we proposed a novel CCA method in which spatial filters are estimated using training data only. This is achieved by using observed EEG training data and their SSVEP components as the two inputs of CCA and the objective function is optimized by averaging multiple training trials. In this case, we proved in theory that the two spatial filters estimated by the CCA are equivalent, and that the CCA and TRCA are also equivalent under certain hypotheses. A benchmark SSVEP data set from 35 subjects was used to compare the performance of the two algorithms according to different lengths of data, numbers of channels and numbers of training trials. In addition, the CCA was also compared with power spectral density analysis (PSDA). The experimental results suggest that the CCA is equivalent to TRCA if the signal-to-noise ratio of training data is high enough; otherwise, the CCA outperforms TRCA in terms of classification accuracy. The CCA is much faster than PSDA in detecting time of targets. The robustness of the training data-driven CCA to noise gives it greater potential in practical applications.

A Brain Computer Interface based Communication System using SSVEP and EOG

This study aims to design a high performance communication system using steady state visual evoked potential (SSVEP) and electro-oculogram (EOG) signal. The proposed keyboard system consists of twenty three randomly chosen characters and they were indexed into different flickering visual stimuli designed using three SSVEP stimulus frequencies. The same frequency valued visual stimuli were used many times (more than once) in the keyboard layout. In this approach an oddball paradigm is introduced and it randomly highlights three/two unique frequency valued visual stimuli at a time. The system identifies a desired target by user eye blink in accordance to the oddball paradigm and recognized SSVEP frequency value. The signal to noise ratio (SNR) of the real time SSVEP electroencephalogram signal gets increased by the introduction of oddball paradigm in the SSVEP based keyboard system. The paradigm will increase the subject’s attention and concentration on the flickering target stimulus. The real time SSVEP data were classified using extended multivariate synchronization index (EMSI) method. The thresholding method is used for single blink detection. The offline and online experiments were performed by all the subjects for evaluating the proposed communication system. The average classification accuracy and information transfer rate of the system are 96.73 % and 76.02 bits/min respectively.

A Single-Channel SSVEP-Based BCI Speller Using Deep Learning

This paper aims to develop a speller system based on a bipolar single-channel electroencephalogram with sufficient accuracy. The proposed system consists of a custom-designed headset, a new virtual keyboard with 58 characters, special symbols, and digits, and a five-target steady-state visual-evoked potential (SSVEP)-based brain–computer interface (BCI) utilizing one-dimensional convolutional neural network (1-D CNN) for SSVEP frequency detection. The deep learning model is implemented and trained under the training mode before being applied in the operation mode of the system. To validate the proposed model, we acquire the training dataset with numerous testing conditions, including different frequency resolutions of the feature and different time-window lengths of analysis. Two types of features based on the frequency domain are investigated to compare their performances in terms of classification accuracy of the model. The experimental results from eight subjects shows that on average, the proposed model can classify five-class SSVEP data with a high accuracy of 99.2%. The proposed BCI is then employed in an online experiment of spelling the word “SPELLER” using 2-s time window. Consequently, the system achieves an average accuracy of 97.4% and an information transfer rate of 49 ± 7.7 bpm, showing the practicality and feasibility of implementing a reliable single-channel SSVEP-based speller utilizing 1-D CNN.

Comparison of SSVEP Signal Classification Techniques Using SVM and ANN Models for BCI Applications

In recent years, Brain Computer Interface (BCI) systems based on Steady-State Visual Evoked Potential (SSVEP) have received much attentions. This study tries to develop a classifier, which can provide higher classification accuracy for multiclass SSVEP data. Four different flickering frequencies in low frequency region were used to elicit the SSVEPs and were displayed on a Liquid Crystal Display (LCD) monitor using LabVIEW. The Electroencephalogram (EEG) signals recorded from the occipital region were first segmented into 1 second window and features were extracted using Fast Fourier Transform (FFT). One-Against-All (OAA), a popular strategy for multiclass Support Vector Machines (SVM) is compared with Artificial Neural Network (ANN) models on the basis of SSVEP classifier accuracies. OAA SVM classifier had got an average accuracy of 88.55% for SSVEP classification over 10 subjects. Based on this study, it is found that for SSVEP classification OAA -SVM classifier can provide better results

Prediction of SSVEP-based BCI performance by the resting-state EEG network

Objective. The prediction of brain–computer interface (BCI) performance is a significant topic in the BCI field. Some researches have demonstrated that resting-state data are promising candidates to achieve the goal. However, so far the relationships between the resting-state networks and the steady-state visual evoked potential (SSVEP)-based BCI have not been investigated. In this paper, we investigate the possible relationships between the SSVEP responses, the classification accuracy of five stimulus frequencies and the closed-eye resting-state network topology. Approach. The resting-state functional connectivity networks of the corresponding five stimulus frequencies were created by coherence, and then three network topology measures—the mean functional connectivity, the clustering coefficient and the characteristic path length of each network—were calculated. In addition, canonical correlation analysis was used to perform frequency recognition with the SSVEP data. Main results. Interestingly, we found that SSVEPs of each frequency were negatively correlated with the mean functional connectivity and clustering coefficient, but positively correlated with characteristic path length. Each of the averaged network topology measures across the frequencies showed the same relationship with the SSVEPs averaged across frequencies between the subjects. Furthermore, our results also demonstrated that the classification accuracy can be predicted by three averaged network measures and their combination can further improve the prediction performance. Significance. These findings indicate that the SSVEP responses and performance are predictable using the information at the resting-state, which may be instructive in both SSVEP-aided cognition studies and SSVEP-based BCI applications.