Group-member selection for RSVP-based collaborative brain-computer interfaces

Objective

The brain-computer interface (BCI) systems based on rapid serial visual presentation (RSVP) have been widely utilized for the detection of target and non-target images. Collaborative brain-computer interface (cBCI) effectively fuses electroencephalogram (EEG) data from multiple users to overcome the limitations of low single-user performance in single-trial event-related potential (ERP) detection in RSVP-based BCI systems. In a multi-user cBCI system, a superior group mode may lead to better collaborative performance and lower system cost. However, the key factors that enhance the collaboration capabilities of multiple users and how to further use these factors to optimize group mode remain unclear.

Approach

This study proposed a group-member selection strategy to optimize the group mode and improve the system performance for RSVP-based cBCI. In contrast to the conventional grouping of collaborators at random, the group-member selection strategy enabled pairing each user with a better collaborator and allowed tasks to be done with fewer collaborators. Initially, we introduced the maximum individual capability and maximum collaborative capability (MIMC) to select optimal pairs, improving the system classification performance. The sequential forward floating selection (SFFS) combined with MIMC then selected a sub-group, aiming to reduce the hardware and labor expenses in the cBCI system. Moreover, the hierarchical discriminant component analysis (HDCA) was used as a classifier for within-session conditions, and the Euclidean space data alignment (EA) was used to overcome the problem of inter-trial variability for cross-session analysis.

Main results

In this paper, we verified the effectiveness of the proposed group-member selection strategy on a public RSVP-based cBCI dataset. For the two-user matching task, the proposed MIMC had a significantly higher AUC and TPR and lower FPR than the common random grouping mode and the potential group-member selection method. Moreover, the SFFS with MIMC enabled a trade-off between maintaining performance and reducing the number of system users.

Significance

The results showed that our proposed MIMC effectively optimized the group mode, enhanced the classification performance in the two-user matching task, and could reduce the redundant information by selecting the sub-group in the RSVP-based multi-user cBCI systems.

An auto-segmented multi-time window dual-scale neural network for brain-computer interfaces based on event-related potentials

Objective.Event-related potentials (ERPs) are cerebral responses to cognitive processes, also referred to as cognitive potentials. Accurately decoding ERPs can help to advance research on brain-computer interfaces (BCIs). The spatial pattern of ERP varies with time. In recent years, convolutional neural networks (CNNs) have shown promising results in electroencephalography (EEG) classification, specifically for ERP-based BCIs.Approach.This study proposes an auto-segmented multi-time window dual-scale neural network (AWDSNet). The combination of a multi-window design and a lightweight base network gives AWDSNet good performance at an acceptable cost of computing. For each individual, we create a time window set by calculating the correlation of signedR-squared values, which enables us to determine the length and number of windows automatically. The signal data are segmented based on the obtained window sets in sub-plus-global mode. Then, the multi-window data are fed into a dual-scale CNN model, where the sizes of the convolution kernels are determined by the window sizes. The use of dual-scale spatiotemporal convolution focuses on feature details while also having a large enough receptive length, and the grouping parallelism undermines the increase in the number of parameters that come with dual scaling.Main results.We evaluated the performance of AWDSNet on a public dataset and a self-collected dataset. A comparison was made with four popular methods including EEGNet, DeepConvNet, EEG-Inception, and PPNN. The experimental results show that AWDSNet has excellent classification performance with acceptable computational complexity.Significance.These results indicate that AWDSNet has great potential for applications in ERP decoding.

Multiband task related components enhance rapid cognition decoding for both small and similar objects

The cortically-coupled target recognition system based on rapid serial visual presentation (RSVP) has a wide range of applications in brain computer interface (BCI) fields such as medical and military. However, in the complex natural environment backgrounds, the identification of event-related potentials (ERP) of both small and similar objects that are quickly presented is a research challenge. Therefore, we designed corresponding experimental paradigms and proposed a multi-band task related components matching (MTRCM) method to improve the rapid cognitive decoding of both small and similar objects. We compared the areas under the receiver operating characteristic curve (AUC) between MTRCM and other 9 methods under different numbers of training sample using RSVP-ERP data from 50 subjects. The results showed that MTRCM maintained an overall superiority and achieved the highest average AUC (0.6562 ± 0.0091). We also optimized the frequency band and the time parameters of the method. The verification on public data sets further showed the necessity of designing MTRCM method. The MTRCM method provides a new approach for neural decoding of both small and similar RSVP objects, which is conducive to promote the further development of RSVP-BCI.

A Cascade xDAWN EEGNet Structure for Unified Visual-Evoked Related Potential Detection

Visual-based brain-computer interface (BCI) enables people to communicate with others by spelling words from the brain and helps professionals recognize targets in large numbers of images. P300 signals evoked by different types of stimuli, such as words or images, may vary significantly in terms of both amplitude and latency. A unified approach is required to detect variable P300 signals, which facilitates BCI applications, as well as deepens the understanding of the P300 generation mechanism. In this study, our proposed approach involves a cascade network structure that combines xDAWN and classical EEGNet techniques. This network is designed to classify target and non-target stimuli in both P300 speller and rapid serial visual presentation (RSVP) paradigms. The proposed approach is capable of recognizing more symbols with fewer repetitions (up to 5 rounds) compared to other models while possessing a better information transfer rate (ITR) as demonstrated on Dataset II (17.22 bits/min in the second repetition round) of BCI Competition III. Additionally, our approach has the highest unweighted average recall (UAR) performance for both 5 Hz ( 0.8134±0.0259 ) and 20 Hz ( 0.6527±0.0321 ) RSVP. The results show that the cascade network structure has better performance between both the P300 Speller and RSVP paradigms, manifesting that such a cascade structure is robust enough for dealing with P300-related signals (source code is available at https://github.com/embneural/Cascade-xDAWN-EEGNet-for-ERP-Detection).

Semantic image sorting method for RSVP presentation

Objective.The rapid serial visual presentation (RSVP) paradigm, which is based on the electroencephalogram (EEG) technology, is an effective approach for object detection. It aims to detect the event-related potentials (ERP) components evoked by target images for rapid identification. However, the object detection performance within this paradigm is affected by the visual disparity between adjacent images in a sequence. Currently, there is no objective metric to quantify this visual difference. Consequently, a reliable image sorting method is required to ensure the generation of a smooth sequence for effective presentation.Approach. In this paper, we propose a novel semantic image sorting method for sorting RSVP sequences, which aims at generating sequences that are perceptually smoother in terms of the human visual experience.Main results. We conducted a comparative analysis between our method and two existing methods for generating RSVP sequences using both qualitative and quantitative assessments. A qualitative evaluation revealed that the sequences generated by our method were smoother in subjective vision and were more effective in evoking stronger ERP components than those generated by the other two methods. Quantitatively, our method generated semantically smoother sequences than the other two methods. Furthermore, we employed four advanced approaches to classify single-trial EEG signals evoked by each of the three methods. The classification results of the EEG signals evoked by our method were superior to those of the other two methods.Significance. In summary, the results indicate that the proposed method can significantly enhance the object detection performance in RSVP-based sequences.

Attention-ProNet: A Prototype Network with Hybrid Attention Mechanisms Applied to Zero Calibration in Rapid Serial Visual Presentation-Based Brain-Computer Interface

The rapid serial visual presentation-based brain-computer interface (RSVP-BCI) system achieves the recognition of target images by extracting event-related potential (ERP) features from electroencephalogram (EEG) signals and then building target classification models. Currently, how to reduce the training and calibration time for classification models across different subjects is a crucial issue in the practical application of RSVP. To address this issue, a zero-calibration (ZC) method termed Attention-ProNet, which involves meta-learning with a prototype network integrating multiple attention mechanisms, was proposed in this study. In particular, multiscale attention mechanisms were used for efficient EEG feature extraction. Furthermore, a hybrid attention mechanism was introduced to enhance model generalization, and attempts were made to incorporate suitable data augmentation and channel selection methods to develop an innovative and high-performance ZC RSVP-BCI decoding model algorithm. The experimental results demonstrated that our method achieved a balance accuracy (BA) of 86.33% in the decoding task for new subjects. Moreover, appropriate channel selection and data augmentation methods further enhanced the performance of the network by affording an additional 2.3% increase in BA. The model generated by the meta-learning prototype network Attention-ProNet, which incorporates multiple attention mechanisms, allows for the efficient and accurate decoding of new subjects without the need for recalibration or retraining.

Possibilistic distribution distance metric: a robust domain adaptation learning method

The affective Brain-Computer Interface (aBCI) systems, which achieve predictions for individual subjects through training on multiple subjects, often cannot achieve satisfactory results due to the differences in Electroencephalogram (EEG) patterns between subjects. One tried to use Subject-specific classifiers, but there was a lack of sufficient labeled data. To solve this problem, Domain Adaptation (DA) has recently received widespread attention in the field of EEG-based emotion recognition. Domain adaptation (DA) learning aims to solve the problem of inconsistent distributions between training and test datasets and has received extensive attention. Most existing methods use Maximum Mean Discrepancy (MMD) or its variants to minimize the problem of domain distribution inconsistency. However, noisy data in the domain can lead to significant drift in domain means, which can affect the adaptability performance of learning methods based on MMD and its variants to some extent. Therefore, we propose a robust domain adaptation learning method with possibilistic distribution distance measure. Firstly, the traditional MMD criterion is transformed into a novel possibilistic clustering model to weaken the influence of noisy data, thereby constructing a robust possibilistic distribution distance metric (P-DDM) criterion. Then the robust effectiveness of domain distribution alignment is further improved by a fuzzy entropy regularization term. The proposed P-DDM is in theory proved which be an upper bound of the traditional distribution distance measure method MMD criterion under certain conditions. Therefore, minimizing P-DDM can effectively optimize the MMD objective. Secondly, based on the P-DDM criterion, a robust domain adaptation classifier based on P-DDM (C-PDDM) is proposed, which adopts the Laplacian matrix to preserve the geometric consistency of instances in the source domain and target domain for improving the label propagation performance. At the same time, by maximizing the use of source domain discriminative information to minimize domain discrimination error, the generalization performance of the learning model is further improved. Finally, a large number of experiments and analyses on multiple EEG datasets (i.e., SEED and SEED-IV) show that the proposed method has superior or comparable robustness performance (i.e., has increased by around 10%) in most cases.

Dual stimuli interface with logical division using local move stimuli

Improving information transfer rate is a key to prompt the speed of outputting instructions of the event-related potential-based brain-computer interface. Our previous study designed a dual-stimuli interface that simultaneously presents two types of different stimuli to improve the speed. While, adding more stimuli into this interface makes subject easily affected by "flanker effect" that decreases the accuracy of recognizing intention. To achieve high recognition accuracy with many stimuli, this study proposes a dual stimuli interface based on whole flash and local move (DS-WL) and two rules of stimulus arrangement to induce the brain signals. Twenty subjects participated in the experiment, and their signals are recognized by a back propagation neural network classifier. The local move induces larger and later signals of targets to help discriminate the two kinds of stimuli; the rules reduce the N200 and P300 amplitudes of non-target, which improves accuracy. This study demonstrates that the DS-WL is a useful way to shorten the instruction output cycle and speed up the instructions outputting by local move and rules.

ETucker: a constrained tensor decomposition for single trial ERP extraction

Objective.In this paper, we propose a new tensor decomposition to extract event-related potentials (ERP) by adding a physiologically meaningful constraint to the Tucker decomposition.Approach.We analyze the performance of the proposed model and compare it with Tucker decomposition by synthesizing a dataset. The simulated dataset is generated using a 12th-order autoregressive model in combination with independent component analysis (ICA) on real no-task electroencephalogram (EEG) recordings. The dataset is manipulated to contain the P300 ERP component and to cover different SNR conditions, ranging from 0 to -30 dB, to simulate the presence of the P300 component in extremely noisy recordings. Furthermore, in order to assess the practicality of the proposed methodology in real-world scenarios, we utilized the brain-computer interface (BCI) competition III-dataset II.Main results.Our primary results demonstrate the superior performance of our approach compared to conventional methods commonly employed for single-trial estimation. Additionally, our method outperformed both Tucker decomposition and non-negative Tucker decomposition in the synthesized dataset. Furthermore, the results obtained from real-world data exhibited meaningful performance and provided insightful interpretations for the extracted P300 component.Significance.The findings suggest that the proposed decomposition is eminently capable of extracting the target P300 component's waveform, including latency and amplitude as well as its spatial location, using single-trial EEG recordings.

Assembling global and local spatial-temporal filters to extract discriminant information of EEG in RSVP task

Objective.Brain-computer interface (BCI) system has developed rapidly in the past decade. And rapid serial visual presentation (RSVP) is an important BCI paradigm to detect the targets in high-speed image streams. For decoding electroencephalography (EEG) in RSVP task, the ensemble-model methods have better performance than the single-model ones.Approach.This study proposed a method based on ensemble learning to extract discriminant information of EEG. An extreme gradient boosting framework was utilized to sequentially generate the sub models, including one global spatial-temporal filter and a group of local ones. EEG was reshaped into a three-dimensional form by remapping the electrode dimension into a 2D array to learn the spatial-temporal features from real local space.Main results.A benchmark RSVP EEG dataset was utilized to evaluate the performance of the proposed method, where EEG data of 63 subjects were analyzed. Compared with several state-of-the-art methods, the spatial-temporal patterns of proposed method were more consistent with P300, and the proposed method can provide significantly better classification performance.Significance.The ensemble model in this study was end-to-end optimized, which can avoid error accumulation. The sub models optimized by gradient boosting theory can extract discriminant information complementarily and non-redundantly.

Novel electrotactile brain-computer interface with somatosensory event-related potential based control

Objective

A brain computer interface (BCI) allows users to control external devices using non-invasive brain recordings, such as electroencephalography (EEG). We developed and tested a novel electrotactile BCI prototype based on somatosensory event-related potentials (sERP) as control signals, paired with a tactile attention task as a control paradigm.

Approach

A novel electrotactile BCI comprises commercial EEG device, an electrical stimulator and custom software for EEG recordings, electrical stimulation control, synchronization between devices, signal processing, feature extraction, selection, and classification. We tested a novel BCI control paradigm based on tactile attention on a sensation at a target stimulation location on the forearm. Tactile stimuli were electrical pulses delivered at two proximal locations on the user's forearm for stimulating branches of radial and median nerves, with equal probability of the target and distractor stimuli occurrence, unlike in any other ERP-based BCI design. We proposed a compact electrical stimulation electrodes configuration for delivering electrotactile stimuli (target and distractor) using 2 stimulation channels and 3 stimulation electrodes. We tested the feasibility of a single EEG channel BCI control, to determine pseudo-online BCI performance, in ten healthy subjects. For optimizing the BCI performance we compared the results for two classifiers, sERP averaging approaches, and novel dedicated feature extraction/selection methods via cross-validation procedures.

Main results

We achieved a single EEG channel BCI classification accuracy in the range of 75.1 to 88.1% for all subjects. We have established an optimal combination of: single trial averaging to obtain sERP, feature extraction/selection methods and classification approach.

Significance

The obtained results demonstrate that a novel electrotactile BCI paradigm with equal probability of attended (target) and unattended (distractor) stimuli and proximal stimulation sites is feasible. This method may be used to drive restorative BCIs for sensory retraining in stroke or brain injury, or assistive BCIs for communication in severely disabled users.

Single-trial P300 classification algorithm based on centralized multi-person data fusion CNN

Introduction

Currently, it is still a challenge to detect single-trial P300 from electroencephalography (EEG) signals. In this paper, to address the typical problems faced by existing single-trial P300 classification, such as complex, time-consuming and low accuracy processes, a single-trial P300 classification algorithm based on multiplayer data fusion convolutional neural network (CNN) is proposed to construct a centralized collaborative brain-computer interfaces (cBCI) for fast and highly accurate classification of P300 EEG signals.

Methods

In this paper, two multi-person data fusion methods (parallel data fusion and serial data fusion) are used in the data pre-processing stage to fuse multi-person EEG information stimulated by the same task instructions, and then the fused data is fed as input to the CNN for classification. In building the CNN network for single-trial P300 classification, the Conv layer was first used to extract the features of single-trial P300, and then the Maxpooling layer was used to connect the Flatten layer for secondary feature extraction and dimensionality reduction, thereby simplifying the computation. Finally batch normalisation is used to train small batches of data in order to better generalize the network and speed up single-trial P300 signal classification.

Results

In this paper, the above new algorithms were tested on the Kaggle dataset and the Brain-Computer Interface (BCI) Competition III dataset, and by analyzing the P300 waveform features and EEG topography and the four standard evaluation metrics, namely Accuracy, Precision, Recall and F1-score,it was demonstrated that the single-trial P300 classification algorithm after two multi-person data fusion CNNs significantly outperformed other classification algorithms.

Discussion

The results show that the single-trial P300 classification algorithm after two multi-person data fusion CNNs significantly outperformed the single-person model, and that the single-trial P300 classification algorithm with two multi-person data fusion CNNs involves smaller models, fewer training parameters, higher classification accuracy and improves the overall P300-cBCI classification rate and actual performance more effectively with a small amount of sample information compared to other algorithms.

Introducing block-Toeplitz covariance matrices to remaster linear discriminant analysis for event-related potential brain-computer interfaces

Objective.Covariance matrices of noisy multichannel electroencephalogram (EEG) time series data provide essential information for the decoding of brain signals using machine learning methods. However, small datasets and high dimensionality make it hard to estimate these matrices. In brain-computer interfaces (BCI) based on event-related potentials (ERP) and a linear discriminant analysis (LDA) classifier, the state of the art covariance estimation uses shrinkage regularization. As this is a general covariance regularization approach, we aim at improving LDA further by better exploiting the domain-specific characteristics of the EEG to regularize the covariance estimates.Approach.We propose to enforce a block-Toeplitz structure for the covariance matrix of the LDA, which implements an assumption of signal stationarity in short time windows for each channel.Main results.An offline re-analysis of data collected from 213 subjects under 13 different event-related potential BCI protocols showed a significantly increased binary classification performance of this 'ToeplitzLDA' compared to shrinkage regularized LDA (up to 6 AUC points,p < 0.001) and Riemannian classification approaches (up to 2 AUC points,p < 0.001). In an unsupervised visual speller application, this improvement would translate to a relative reduction of spelling errors by 81% on average for 25 subjects. Additionally, aside from lower memory and reduced time complexity for LDA training, ToeplitzLDA proves to be robust against drastic increases of the number of temporal features.Significance.The proposed covariance estimation allows BCI researchers to improve classification rates and reduce calibration times of BCI protocols using event-related potentials and thus support the usability of corresponding applications. Its lower computational and memory needs could make it a valuable algorithm especially for mobile BCIs.

A Novel Deep Learning Method Based on an Overlapping Time Window Strategy for Brain-Computer Interface-Based Stroke Rehabilitation

Globally, stroke is a leading cause of death and disability. The classification of motor intentions using brain activity is an important task in the rehabilitation of stroke patients using brain-computer interfaces (BCIs). This paper presents a new method for model training in EEG-based BCI rehabilitation by using overlapping time windows. For this aim, three different models, a convolutional neural network (CNN), graph isomorphism network (GIN), and long short-term memory (LSTM), are used for performing the classification task of motor attempt (MA). We conducted several experiments with different time window lengths, and the results showed that the deep learning approach based on overlapping time windows achieved improvements in classification accuracy, with the LSTM combined vote-counting strategy (VS) having achieved the highest average classification accuracy of 90.3% when the window size was 70. The results verified that the overlapping time window strategy is useful for increasing the efficiency of BCI rehabilitation.

A novel brain-computer interface based on audio-assisted visual evoked EEG and spatial-temporal attention CNN

Objective

Brain-computer interface (BCI) can translate intentions directly into instructions and greatly improve the interaction experience for disabled people or some specific interactive applications. To improve the efficiency of BCI, the objective of this study is to explore the feasibility of an audio-assisted visual BCI speller and a deep learning-based single-trial event related potentials (ERP) decoding strategy.

Approach

In this study, a two-stage BCI speller combining the motion-onset visual evoked potential (mVEP) and semantically congruent audio evoked ERP was designed to output the target characters. In the first stage, the different group of characters were presented in the different locations of visual field simultaneously and the stimuli were coded to the mVEP based on a new space division multiple access scheme. And then, the target character can be output based on the audio-assisted mVEP in the second stage. Meanwhile, a spatial-temporal attention-based convolutional neural network (STA-CNN) was proposed to recognize the single-trial ERP components. The CNN can learn 2-dimentional features including the spatial information of different activated channels and time dependence among ERP components. In addition, the STA mechanism can enhance the discriminative event-related features by adaptively learning probability weights.

Main results

The performance of the proposed two-stage audio-assisted visual BCI paradigm and STA-CNN model was evaluated using the Electroencephalogram (EEG) recorded from 10 subjects. The average classification accuracy of proposed STA-CNN can reach 59.6 and 77.7% for the first and second stages, which were always significantly higher than those of the comparison methods (p < 0.05).

Significance

The proposed two-stage audio-assisted visual paradigm showed a great potential to be used to BCI speller. Moreover, through the analysis of the attention weights from time sequence and spatial topographies, it was proved that STA-CNN could effectively extract interpretable spatiotemporal EEG features.

A New PLV-Spatial Filtering to Improve the Classification Performance in BCI Systems

OBJECTIVE

The performance of an EEG-based brain-computer interface (BCI) system is highly dependent on signal preprocessing. This manuscript presents a filtering method to improve the feature classification algorithms typically used in BCI.

METHODS

A graph Laplacian quadratic form using the Phase Locking Value (PLV) is applied to generate a new filtered signal in the preprocessing stage.

RESULTS

The accuracy of the classification algorithms improved significantly (up to 27.18% in the BCI Competition IV dataset, and up to 42.56% with records made with an Emotiv EPOC+). In addition, the proposed filtering algorithm has similar or better results when compared with the Filter Bank Common Spatial Pattern (FBCSP), which has disadvantages in a multiclass classification.

CONCLUSION

This paper shows how our PLV-based filtering between EEG channels could improve the performance of a BCI.

Event-Related Potential-Based Brain-Computer Interface Using the Thai Vowels' and Numerals' Auditory Stimulus Pattern

Herein, we developed an auditory stimulus pattern for an event-related potential (ERP)-based brain-computer interface (BCI) system to improve control and communication in quadriplegia with visual impairment. Auditory stimulus paradigms for multicommand electroencephalogram (EEG)-based BCIs and audio stimulus patterns were examined. With the proposed auditory stimulation, using the selected Thai vowel, similar to the English vowel, and Thai numeral sounds, as simple target recognition, we explored the ERPs' response and classification efficiency from the suggested EEG channels. We also investigated the use of single and multi-loudspeakers for auditory stimuli. Four commands were created using the proposed paradigm. The experimental paradigm was designed to observe ERP responses and verify the proposed auditory stimulus pattern. The conventional classification method produced four commands using the proposed auditory stimulus pattern. The results established that the proposed auditory stimulation with 20 to 30 trials of stream stimuli could produce a prominent ERP response from Pz channels. The vowel stimuli could achieve higher accuracy than the proposed numeral stimuli for two auditory stimuli intervals (100 and 250 ms). Additionally, multi-loudspeaker patterns through vowel and numeral sound stimulation provided an accuracy greater than 85% of the average accuracy. Thus, the proposed auditory stimulation patterns can be implemented as a real-time BCI system to aid in the daily activities of quadratic patients with visual and tactile impairments. In future, practical use of the auditory ERP-based BCI system will be demonstrated and verified in an actual scenario.

Bidirectional siamese correlation analysis method for enhancing the detection of SSVEPs

OBJECTIVE

Steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs) have attracted increasing attention due to their high information transfer rate. To improve the performance of SSVEP detection, we propose a bidirectional Siamese correlation analysis (bi-SiamCA) model.

APPROACH

In this model, an LSTM-based Siamese architecture is designed to measure the similarity between the SSVEP signal and the template in each frequency and obtain the probability that the SSVEP signal belongs to each frequency. Additionally, a maximize agreement module with a designed contrastive loss is adopted in the Siamese architecture to increase the similarity between the SSVEP signal and the reference signal in the same frequency. Moreover, a two-way signal processing mechanism is built to effectively integrate complementary information from two temporal directions of the input signals. Our model uses raw SSVEPs as inputs and can be trained end-to-end.

MAIN RESULTS

Experimental results on a 40-class dataset and a 12-class dataset indicate that bi-SiamCA can significantly improve the classification accuracy compared with the prominent traditional and deep learning methods, especially under short data lengths. Feature visualizations show that the similarity between the SSVEP signal and the reference signal in the same frequency gradually improved in our model.

CONCLUSION

The proposed bi-SiamCA model enhances the performance of SSVEP detection and outperforms the compared methods.

SIGNIFICANCE

Due to its high decoding accuracy under short signals, our approach has great potential to implement a high-speed SSVEP-based BCI.

Behavioral Analysis of EEG Signals in Loss-Gain Decision-Making Experiments

Extraction and analysis of the EEG (electroencephalograph) information features generated during behavioral decision-making can provide a better understanding of the state of mind. Previous studies have focused more on the brainwave features after behavioral decision-making. In fact, the EEG before decision-making is more worthy of our attention. In this study, we introduce a new index based on the reaction time of subjects before decision-making, called the Prestimulus Time (PT), which have important reference value for the study of cognitive function, neurological diseases, and other fields. In our experiments, we use a wearable EEG feature signal acquisition device and a systematic reward and punishment experiment to obtain the EEG features before and after behavioral decision-making. The experimental results show that the EEG generated after behavioral decision due to loss is more intense than that generated by gain in the medial frontal cortex (MFC). In addition, different characteristics of EEG signals are generated prior to behavioral decisions because people have different expectations of the outcome. It will produce more significant negative-polarity event-related potential (ERP) in the forebrain area when the humans are optimistic about the outcomes.

SAST-GCN: Segmentation Adaptive Spatial Temporal-Graph Convolutional Network for P3-Based Video Target Detection

Detecting video-induced P3 is crucial to building the video target detection system based on the brain-computer interface. However, studies have shown that the brain response patterns corresponding to video-induced P3 are dynamic and determined by the interaction of multiple brain regions. This paper proposes a segmentation adaptive spatial-temporal graph convolutional network (SAST-GCN) for P3-based video target detection. To make full use of the dynamic characteristics of the P3 signal data, the data is segmented according to the processing stages of the video-induced P3, and the brain network connections are constructed correspondingly. Then, the spatial-temporal feature of EEG data is extracted by adaptive spatial-temporal graph convolution to discriminate the target and non-target in the video. Especially, a style-based recalibration module is added to select feature maps with higher contributions and increase the feature extraction ability of the network. The experimental results demonstrate the superiority of our proposed model over the baseline methods. Also, the ablation experiments indicate that the segmentation of data to construct the brain connection can effectively improve the recognition performance by reflecting the dynamic connection relationship between EEG channels more accurately.

An improved EEGNet for single-trial EEG classification in rapid serial visual presentation task

As a new type of brain–computer interface (BCI), the rapid serial visual presentation (RSVP) paradigm has attracted significant attention. The mechanism of RSVP is detecting the P300 component corresponding to the target image to realize fast and correct recognition. This paper proposed an improved EEGNet model to achieve good performance in offline and online data. Specifically, the data were filtered by xDAWN to enhance the signal-to-noise ratio of the electroencephalogram (EEG) signals. The focal loss function was used instead of the cross-entropy loss function to solve the classification problems of unbalanced samples. Additionally, the subject-specific data were fed to the improved EEGNet model to obtain a subject-specific model. We applied the proposed model at the BCI Controlled Robot Contest in World Robot Contest 2021 and won the second place. The average recall rate of the four participants reached 51.56% in triple classification. In the offline data benchmark dataset (64 subjects-RSVP tasks), the average recall rates of groups A and B reached 76.07% and 78.11%, respectively. We provided an alternative method to identify targets based on the RSVP paradigm.

Task-Related Component Analysis Combining Paired Character Decoding for Miniature Asymmetric Visual Evoked Potentials

Brain-computer interface (BCI) technology based on event-related potentials (ERP) of electroencephalography (EEG) is widely used in daily life and medical treatment. However, the research of identifying the miniature and more informative asymmetric visual evoked potentials (aVEPs), which belongs to ERP, needs further exploration. Herein, a task-related component analysis combining paired character decoding (TRCA-PCD) method, which can enhance reproducibility of aVEPs in multiple trials and strengthen the features of different samples, was designed to realize fast decoding of aVEPs. The BCI performance and the influence of repetition times between the TRCA-PCD method, the discriminative canonical pattern matching (DCPM) method and traditional task-related component analysis (TRCA) method were compared using a 32-class aVEPs dataset recorded from 32 subjects. The highest average recognition accuracy and information transfer rate (ITR) of TRCA-PCD after parameter selection were 70.37 ± 2.49% (DCPM: 64.91 ± 2.81%, TRCA: 44.01 ± 3.25%) with the peak value of 97.92% and 28.90 ± 3.83 bits/min (DCPM: 21.29 ± 3.35 bits/min, TRCA: 11.54 ± 2.81 bits/min) with the peak value of 94.55 bits/min respectively. Statistical analysis indicated that the highest average recognition rate could be obtained when the repetition time was six, and the highest ITR could be obtained when the repetition time was one. Overall, the results verified the effectiveness and superiority of TRCA-PCD in recognition of aVEPs and provided a reference for parameter selection. Therefore, the TRCA-PCD method can promote the further application of aVEPs in the BCI speller field.

Decoupling Representation Learning for Imbalanced Electroencephalography Classification in Rapid Serial Visual Presentation Task

OBJECTIVE

The class imbalance problem considerably restricts the performance of electroencephalography (EEG) classification in the rapid serial visual presentation (RSVP) task. Existing solutions typically employ re-balancing strategies (e.g. re-weighting and re-sampling) to alleviate the impact of class imbalance, which enhances the classifier learning of deep networks but unexpectedly damages the representative ability of the learned deep features as original distributions become distorted.

APPROACH

In this study, a novel decoupling representation learning (DRL) model, has been proposed that separates the representation learning and classification processes to capture the discriminative feature of imbalanced RSVP EEG data while classifying it accurately. The representation learning process is responsible for learning universal patterns for the classification of all samples, while the classifier determines a better bounding for the target and non-target classes. Specifically, the representation learning process adopts a dual-branch architecture, which minimizes the contrastive loss to regularize the representation space. In addition, to learn more discriminative information from RSVP EEG data, a novel multi-granular information (MGI) based extractor is designed to extract spatial-temporal information. Considering the class re-balancing strategies can significantly promote classifier learning, the classifier was trained with rebalanced EEG data while freezing the parameters of the representation learning process.

MAIN RESULTS

To evaluate the proposed method, experiments were conducted on two public datasets and one self-conducted dataset. The results demonstrate that the proposed DRL can achieve state-of-the-art performance for EEG classification in the RSVP task.

SIGNIFICANCE

This is the first study to focus on the class imbalance problem and propose a generic solution in the RSVP task. Furthermore, multi-granular data was explored to extract more complementary spatial-temporal information. The code is open-source and available at https://github.com/Tammie-Li/DRL.

Effects of inter-stimulus intervals on concurrent P300 and SSVEP features for hybrid Brain-computer interfaces

BACKGROUND

Recently, we have implemented a high-speed brain-computer interface (BCI) system with a large instruction set using the concurrent P300 and steady-state visual evoked potential (SSVEP) features (also known as hybrid features). However, it remains unclear how to select inter-stimulus interval (ISI) for the proposed BCI system to balance the encoding efficiency and decoding performance.

NEW METHOD

This study developed a 6⁎9 hybrid P300-SSVEP BCI system and investigated a series of ISIs ranged from -175ms to 0ms with a step of 25ms. The influence of ISI on the hybrid features was analyzed from several aspects, including the amplitude of the induced features, classification accuracy, information transfer rate (ITR). Twelve naive subjects were recruited for the experiment.

RESULTS

The results showed the ISI factor had a significant impact on the hybrid features. Specifically, as the values of ISI decreased, the amplitudes of the induced features and accuracies decreased gradually, while the ITRs increased rapidly. It's achieved the highest ITR of 158.50 bits/min when ISI equal to -175ms.

COMPARISON WITH EXISTING METHOD

The optimal ISI in this study achieved superior performance in comparison with the one we used in the previous study.

CONCLUSIONS

The ISI can exert an important influence on the P300-SSVEP BCI system and its optimal value is -175ms in this study, which is significant for developing the high-speed BCI system with larger instruction sets in the future.

Improving Cross-State and Cross-Subject Visual ERP-based BCI with Temporal Modeling and Adversarial Training

Brain-computer interface (BCI) is a useful device for people without relying on peripheral nerves and muscles. However, the performance of the event-related potential (ERP)-based BCI declines when applying it to real environments, especially in cross-state and cross-subject conditions. Here we employ temporal modeling and adversarial training to improve the visual ERP-based BCI under different mental workload states and to alleviate the problems above. The rationality of our method is that the ERP-based BCI is based on electroencephalography (EEG) signals recorded from the scalp's surface, continuously changing with time and somewhat stochastic. In this paper, we propose a hierarchical recurrent network to encode all ERP signals in each repetition at the same time and model them with a temporal manner to predict which visual event elicited an ERP. The hierarchical architecture is a simple yet effective method for organizing recurrent layers in a deep structure to model long sequence signals. Taking a cue from recent advances in adversarial training, we further applied dynamic adversarial perturbations to create adversarial examples to enhance the model performance. We conduct our experiments on one published visual ERP-based BCI task with 15 subjects and 3 different auditory workload states. The results indicate that our hierarchical method can effectively model the long sequence EEG raw data, outperform the baselines on most conditions, including cross-state and cross-subject conditions. Finally, we show how deep learning-based methods with limited EEG data can improve ERP-based BCI with adversarial training. Our code will be released at https://github.com/aispeech-lab/VisBCI.

A Spatio-Temporal Method for Extracting Gamma-Band Features to Enhance Classification in a Rapid Serial Visual Presentation Task

Rapid serial visual presentation (RSVP) is a type of electroencephalogram (EEG) pattern commonly used for target recognition. Besides delta- and theta-band responses already used for classification, RSVP task also evokes gamma-band responses having low amplitude and large individual difference. This paper proposes a filter bank spatio-temporal component analysis (FBSCA) method, extracting spatio-temporal features of the gamma-band responses for the first time, to enhance the RSVP classification performance. Considering the individual difference in time latency and responsive frequency, the proposed FBSCA method decomposes the gamma-band EEG data into sub-components in different time-frequency-space domains and seeks the weight coefficients to optimize the combinations of electrodes, common spatial pattern (CSP) components, time windows and frequency bands. Two state-of-the-art methods, i.e. hierarchical discriminant principal component analysis (HDPCA) and discriminative canonical pattern matching (DCPM), were used for comparison. The performance was evaluated in [Formula: see text] cross validations using a public dataset. Study results showed that the FBSCA method outperformed the other methods regardless of number of training trials. These results suggest that the proposed FBSCA method can enhance the RSVP classification.

Visual field inhomogeneous in brain-computer interfaces based on rapid serial visual presentation

OBJECTIVE

Visual attention is not homogeneous across the visual field, while how to mine the effective EEG characteristics that are sensitive to the inhomogeneous of visual attention and further explore applications such as the performance of brain-computer interface (BCI) are still distressing explorative scientists.

APPROACH

Images were encoded into a rapid serial visual presentation (RSVP) paradigm, and were presented in three visuospatial patterns (central, left/right, upper/lower) at the stimulation frequencies of 10Hz, 15Hz and 20Hz. The comparisons among different visual fields were conducted in the dimensions of subjective behavioral and EEG characteristics. Furthermore, the effective features (e.g. SSVEP, N2pc and P300) that sensitive to visual-field asymmetry were also explored.

RESULTS

The visual fields had significant influences on the performance of RSVP target detection, in which the performance of central was better than that of peripheral visual field, the performance of horizontal meridian was better than that of vertical meridian, the performance of left visual field was better than that of right visual field, and the performance of upper visual field was better than that of lower visual field. Furthermore, stimuli of different visual fields had significant effects on the spatial distributions of EEG, in which N2pc and P300 showed left-right asymmetry in occipital and frontal regions, respectively. In addition, the evidences of SSVEP characteristics indicated that there was obvious overlap of visual fields on the horizontal meridian, but not on the vertical meridian.

SIGNIFICANCE

The conclusions of this study provide insights into the relationship between visual field inhomogeneous and EEG characteristics. In addition, this study has the potential to achieve precise positioning of the target's spatial orientation in RSVP-BCIs.

A Novel Classification Framework Using the Graph Representations of Electroencephalogram for Motor Imagery based Brain-Computer Interface

The motor imagery (MI) based brain-computer interfaces (BCIs) have been proposed as a potential physical rehabilitation technology. However, the low classification accuracy achievable with MI tasks is still a challenge when building effective BCI systems. We propose a novel MI classification model based on measurement of functional connectivity between brain regions and graph theory. Specifically, motifs describing local network structures in the brain are extracted from functional connectivity graphs. A graph embedding model called Ego-CNNs is then used to build a classifier, which can convert the graph from a structural representation to a fixed-dimensional vector for detecting critical structure in the graph. We validate our proposed method on four datasets, and the results show that our proposed method produces high classification accuracies in two-class classification tasks (92.8% for dataset 1, 93.4% for dataset 2, 96.5% for dataset 3, and 80.2% for dataset 4) and multiclass classification tasks (90.33% for dataset 1). Our proposed method achieves a mean Kappa value of 0.88 across nine participants, which is superior to other methods we compared it to. These results indicate that there is a local structural difference in functional connectivity graphs extracted under different motor imagery tasks. Our proposed method has great potential for motor imagery classification in future studies.

Phase Preservation Neural Network for Electroencephalography Classification in Rapid Serial Visual Presentation Task

Neuroscience studies have demonstrated the phase-locked characteristics of some early event-related potential (ERP) components evoked by stimuli. In this study, we propose a phase preservation neural network (PPNN) to learn phase information to improve the Electroencephalography (EEG) classification in a rapid serial visual presentation (RSVP) task. The PPNN consists of three major modules that can produce spatial and temporal representations with the high discriminative ability of the EEG features for classification. We first adopt a stack of dilated temporal convolution layers to extract temporal dynamics while avoiding the loss of phase information. Considering the intrinsic channel dependence of the EEG data, a spatial convolution layer is then applied to obtain the spatial-temporal representation of the input EEG signal. Finally, a fully connected layer is adopted to extract higher-level features for the final classification. The experiments are conducted on two public and one collected EEG datasets from the RSVP task, in which we evaluated the performance and explored the capability of phase preservation of our PPNN model and visualized the extracted features. The experimental results indicate the superiority of the proposed PPNN when compared with previous methods, suggesting the PPNN is a robust model for EEG classification in RSVP task.

Neural Decoding of EEG Signals with Machine Learning: A Systematic Review

Electroencephalography (EEG) is a non-invasive technique used to record the brain's evoked and induced electrical activity from the scalp. Artificial intelligence, particularly machine learning (ML) and deep learning (DL) algorithms, are increasingly being applied to EEG data for pattern analysis, group membership classification, and brain-computer interface purposes. This study aimed to systematically review recent advances in ML and DL supervised models for decoding and classifying EEG signals. Moreover, this article provides a comprehensive review of the state-of-the-art techniques used for EEG signal preprocessing and feature extraction. To this end, several academic databases were searched to explore relevant studies from the year 2000 to the present. Our results showed that the application of ML and DL in both mental workload and motor imagery tasks has received substantial attention in recent years. A total of 75% of DL studies applied convolutional neural networks with various learning algorithms, and 36% of ML studies achieved competitive accuracy by using a support vector machine algorithm. Wavelet transform was found to be the most common feature extraction method used for all types of tasks. We further examined the specific feature extraction methods and end classifier recommendations discovered in this systematic review.

Enhancing transfer performance across datasets for brain-computer interfaces using a combination of alignment strategies and adaptive batch normalization

Objective. Recently, transfer learning (TL) and deep learning (DL) have been introduced to solve intra- and inter-subject variability problems in brain-computer interfaces (BCIs). However, current TL and DL algorithms are usually validated within a single dataset, assuming that data of the test subjects are acquired under the same condition as that of training (source) subjects. This assumption is generally violated in practice because of different acquisition systems and experimental settings across studies and datasets. Thus, the generalization ability of these algorithms needs further validations in a cross-dataset scenario, which is closer to the actual situation. This study compared the transfer performance of pre-trained deep-learning models with different preprocessing strategies in a cross-dataset scenario.Approach. This study used four publicly available motor imagery datasets, each was successively selected as a source dataset, and the others were used as target datasets. EEGNet and ShallowConvNet with four preprocessing strategies, namely channel normalization, trial normalization, Euclidean alignment, and Riemannian alignment, were trained with the source dataset. The transfer performance of pre-trained models was validated on the target datasets. This study also used adaptive batch normalization (AdaBN) for reducing interval covariate shift across datasets. This study compared the transfer performance of using the four preprocessing strategies and that of a baseline approach based on manifold embedded knowledge transfer (MEKT). This study also explored the possibility and performance of fusing MEKT and EEGNet.Main results. The results show that DL models with alignment strategies had significantly better transfer performance than the other two preprocessing strategies. As an unsupervised domain adaptation method, AdaBN could also significantly improve the transfer performance of DL models. The transfer performance of DL models that combined AdaBN and alignment strategies significantly outperformed MEKT. Moreover, the generalizability of EEGNet models that combined AdaBN and alignment strategies could be further improved via the domain adaptation step in MEKT, achieving the best generalization ability among multiple datasets (BNCI2014001: 0.788, PhysionetMI: 0.679, Weibo2014: 0.753, Cho2017: 0.650).Significance. The combination of alignment strategies and AdaBN could easily improve the generalizability of DL models without fine-tuning. This study may provide new insights into the design of transfer neural networks for BCIs by separating source and target batch normalization layers in the domain adaptation process.

Review of brain encoding and decoding mechanisms for EEG-based brain-computer interface

A brain-computer interface (BCI) can connect humans and machines directly and has achieved successful applications in the past few decades. Many new BCI paradigms and algorithms have been developed in recent years. Therefore, it is necessary to review new progress in BCIs. This paper summarizes progress for EEG-based BCIs from the perspective of encoding paradigms and decoding algorithms, which are two key elements of BCI systems. Encoding paradigms are grouped by their underlying neural meachanisms, namely sensory- and motor-related, vision-related, cognition-related and hybrid paradigms. Decoding algorithms are reviewed in four categories, namely decomposition algorithms, Riemannian geometry, deep learning and transfer learning. This review will provide a comprehensive overview of both modern primary paradigms and algorithms, making it helpful for those who are developing BCI systems.

An Open Source-Based BCI Application for Virtual World Tour and Its Usability Evaluation

Brain-computer interfaces can provide a new communication channel and control functions to people with restricted movements. Recent studies have indicated the effectiveness of brain-computer interface (BCI) applications. Various types of applications have been introduced so far in this field, but the number of those available to the public is still insufficient. Thus, there is a need to expand the usability and accessibility of BCI applications. In this study, we introduce a BCI application for users to experience a virtual world tour. This software was built on three open-source environments and is publicly available through the GitHub repository. For a usability test, 10 healthy subjects participated in an electroencephalography (EEG) experiment and evaluated the system through a questionnaire. As a result, all the participants successfully played the BCI application with 96.6% accuracy with 20 blinks from two sessions and gave opinions on its usability (e.g., controllability, completeness, comfort, and enjoyment) through the questionnaire. We believe that this open-source BCI world tour system can be used in both research and entertainment settings and hopefully contribute to open science in the BCI field.

Detection of fixation points using a small visual landmark for brain-computer interfaces

Objective.The speed of visual brain-computer interfaces (v-BCIs) has been greatly improved in recent years. However, the traditional v-BCI paradigms require users to directly gaze at the intensive flickering items, which would cause severe problems such as visual fatigue and excessive visual resource consumption in practical applications. Therefore, it is imperative to develop a user-friendly v-BCI.Approach.According to the retina-cortical relationship, this study developed a novel BCI paradigm to detect the fixation point of eyes using a small visual stimulus that subtended only 0.6° in visual angle and was out of the central visual field. Specifically, the visual stimulus was treated as a landmark to judge the eccentricity and polar angle of the fixation point. Sixteen different fixation points were selected around the visual landmark, i.e. different combinations of two eccentricities (2° and 4°) and eight polar angles (0,π4,π2,3π4,π,5π4,3π2and7π4). Twelve subjects participated in this study, and they were asked to gaze at one out of the 16 points for each trial. A multi-class discriminative canonical pattern matching (Multi-DCPM) algorithm was proposed to decode the user's fixation point.Main results.We found the visual stimulation landmark elicited different spatial event-related potential patterns for different fixation points. Multi-DCPM could achieve an average accuracy of 66.2% with a standard deviation of 15.8% for the classification of the sixteen fixation points, which was significantly higher than traditional algorithms (p⩽0.001). Experimental results of this study demonstrate the feasibility of using a small visual stimulus as a landmark to track the relative position of the fixation point.Significance.The proposed new paradigm provides a potential approach to alleviate the problem of irritating stimuli in v-BCIs, which can broaden the applications of BCIs.

Enhancement for P300-speller classification using multi-window discriminative canonical pattern matching

Objective.P300s are one of the most studied event-related potentials (ERPs), which have been widely used for brain-computer interfaces (BCIs). Thus, fast and accurate recognition of P300s is an important issue for BCI study. Recently, there emerges a lot of novel classification algorithms for P300-speller. Among them, discriminative canonical pattern matching (DCPM) has been proven to work effectively, in which discriminative spatial pattern (DSP) filter can significantly enhance the spatial features of P300s. However, the pattern of ERPs in space varies with time, which was not taken into consideration in the traditional DCPM algorithm.Approach.In this study, we developed an advanced version of DCPM, i.e. multi-window DCPM, which contained a series of time-dependent DSP filters to fine-tune the extraction of spatial ERP features. To verify its effectiveness, 25 subjects were recruited and they were asked to conduct the typical P300-speller experiment.Main results.As a result, multi-window DCPM achieved the character recognition accuracy of 91.84% with only five training characters, which was significantly better than the traditional DCPM algorithm. Furthermore, it was also compared with eight other popular methods, including SWLDA, SKLDA, STDA, BLDA, xDAWN, HDCA, sHDCA and EEGNet. The results showed multi-window DCPM preformed the best, especially using a small calibration dataset. The proposed algorithm was applied to the BCI Controlled Robot Contest of P300 paradigm in 2019 World Robot Conference, and won the first place.Significance.These results demonstrate that multi-window DCPM is a promising method for improving the performance and enhancing the practicability of P300-speller.

Neural mechanism for dynamic distractor processing during video target detection: Insights from time-varying networks in the cerebral cortex

In dynamic video target detection tasks, distractors may suddenly appear due to the dynamicity of the visual scene and the uncertainty of the visual information, strongly influencing participants' attention and target detection performance. Moreover, the neural mechanism that accounts for dynamic distractor processing remains unknown, which makes it difficult to compensate for in EEG-based video target detection. Here, cortical activities with high spatiotemporal resolution were reconstructed using the source localization method. The time-varying networks among important brain regions in different cognitive phases, including information integration, decision-making, and execution, were identified to investigate the neural mechanism of dynamic distractor processing. The experimental results indicated that dynamic distractors could induce a P3-like component. In addition, there was obvious asymmetry between the two hemispheres during video target detection. Specifically, the brain responses induced by dynamic distractors were weak and more concentrated in the left hemisphere during the information integration phase; left superior frontal gyrus activity related to preparation for the presence of distractors was critical, while the attention network and primary visual network, especially in the left visual pathway, were more active for dynamic targets during the decision-making phase. These findings provide guidance for designing an effective EEG-based model for dynamic video target detection.

Target Detection Using Ternary Classification During a Rapid Serial Visual Presentation Task Using Magnetoencephalography Data

Background: The rapid serial visual presentation (RSVP) paradigm is a high-speed paradigm of brain–computer interface (BCI) applications. The target stimuli evoke event-related potential (ERP) activity of odd-ball effect, which can be used to detect the onsets of targets. Thus, the neural control can be produced by identifying the target stimulus. However, the ERPs in single trials vary in latency and length, which makes it difficult to accurately discriminate the targets against their neighbors, the near-non-targets. Thus, it reduces the efficiency of the BCI paradigm.

Methods: To overcome the difficulty of ERP detection against their neighbors, we proposed a simple but novel ternary classification method to train the classifiers. The new method not only distinguished the target against all other samples but also further separated the target, near-non-target, and other, far-non-target samples. To verify the efficiency of the new method, we performed the RSVP experiment. The natural scene pictures with or without pedestrians were used; the ones with pedestrians were used as targets. Magnetoencephalography (MEG) data of 10 subjects were acquired during presentation. The SVM and CNN in EEGNet architecture classifiers were used to detect the onsets of target.

Results: We obtained fairly high target detection scores using SVM and EEGNet classifiers based on MEG data. The proposed ternary classification method showed that the near-non-target samples can be discriminated from others, and the separation significantly increased the ERP detection scores in the EEGNet classifier. Moreover, the visualization of the new method suggested the different underling of SVM and EEGNet classifiers in ERP detection of the RSVP experiment.

Conclusion: In the RSVP experiment, the near-non-target samples contain separable ERP activity. The ERP detection scores can be increased using classifiers of the EEGNet model, by separating the non-target into near- and far-targets based on their delay against targets.

EEG-Inception: A Novel Deep Convolutional Neural Network for Assistive ERP-Based Brain-Computer Interfaces

In recent years, deep-learning models gained attention for electroencephalography (EEG) classification tasks due to their excellent performance and ability to extract complex features from raw data. In particular, convolutional neural networks (CNN) showed adequate results in brain-computer interfaces (BCI) based on different control signals, including event-related potentials (ERP). In this study, we propose a novel CNN, called EEG-Inception, that improves the accuracy and calibration time of assistive ERP-based BCIs. To the best of our knowledge, EEG-Inception is the first model to integrate Inception modules for ERP detection, which combined efficiently with other structures in a light architecture, improved the performance of our approach. The model was validated in a population of 73 subjects, of which 31 present motor disabilities. Results show that EEG-Inception outperforms 5 previous approaches, yielding significant improvements for command decoding accuracy up to 16.0%, 10.7%, 7.2%, 5.7% and 5.1% in comparison to rLDA, xDAWN + Riemannian geometry, CNN-BLSTM, DeepConvNet and EEGNet, respectively. Moreover, EEG-Inception requires very few calibration trials to achieve state-of-the-art performances taking advantage of a novel training strategy that combines cross-subject transfer learning and fine-tuning to increase the feasibility of this approach for practical use in assistive applications.

Performance Analysis With Different Types of Visual Stimuli in a BCI-Based Speller Under an RSVP Paradigm

Brain-Computer Interface (BCI) systems enable an alternative communication channel for severely-motor disabled patients to interact with their environment using no muscular movements. In recent years, the importance of research into non-gaze dependent brain-computer interface paradigms has been increasing, in contrast to the most frequently studied BCI-based speller paradigm (i.e., row-column presentation, RCP). Several visual modifications that have already been validated under the RCP paradigm for communication purposes have not been validated under the most extended non-gaze dependent rapid serial visual presentation (RSVP) paradigm. Thus, in the present study, three different sets of stimuli were assessed under RSVP, with the following communication features: white letters (WL), famous faces (FF), neutral pictures (NP). Eleven healthy subjects participated in this experiment, in which the subjects had to go through a calibration phase, an online phase and, finally, a subjective questionnaire completion phase. The results showed that the FF and NP stimuli promoted better performance in the calibration and online phases, being slightly better in the FF paradigm. Regarding the subjective questionnaires, again both FF and NP were preferred by the participants in contrast to the WL stimuli, but this time the NP stimuli scored slightly higher. These findings suggest that the use of FF and NP for RSVP-based spellers could be beneficial to increase information transfer rate in comparison to the most frequently used letter-based stimuli and could represent a promising communication system for individuals with altered ocular-motor function.

A Spelling Paradigm With an Added Red Dot Improved the P300 Speller System Performance

The traditional P300 speller system uses the flashing row or column spelling paradigm. However, the classification accuracy and information transfer rate of the P300 speller are not adequate for real-world application. To improve the performance of the P300 speller, we devised a new spelling paradigm in which the flashing row or column of a virtual character matrix is covered by a translucent green circle with a red dot in either the upper or lower half (GC-RD spelling paradigm). We compared the event-related potential (ERP) waveforms with a control paradigm (GC spelling paradigm), in which the flashing row or column of a virtual character matrix was covered by a translucent green circle only. Our experimental results showed that the amplitude of P3a at the parietal area and P3b at the frontal–central–parietal areas evoked by the GC-RD paradigm were significantly greater than those induced by the GC paradigm. Higher classification accuracy and information transmission rates were also obtained in the GC-RD system. Our results indicated that the added red dots increased attention and visuospatial information, resulting in an amplitude increase in both P3a and P3b, thereby improving the performance of the P300 speller system.

Single-Option P300-BCI Performance Is Affected by Visual Stimulation Conditions

The P300 paradigm is one of the most promising techniques for its robustness and reliability in Brain-Computer Interface (BCI) applications, but it is not exempt from shortcomings. The present work studied single-trial classification effectiveness in distinguishing between target and non-target responses considering two conditions of visual stimulation and the variation of the number of symbols presented to the user in a single-option visual frame. In addition, we also investigated the relationship between the classification results of target and non-target events when training and testing the machine-learning model with datasets containing different stimulation conditions and different number of symbols. To this end, we designed a P300 experimental protocol considering, as conditions of stimulation: the color highlighting or the superimposing of a cartoon face and from four to nine options. These experiments were carried out with 19 healthy subjects in 3 sessions. The results showed that the Event-Related Potentials (ERP) responses and the classification accuracy are stronger with cartoon faces as stimulus type and similar irrespective of the amount of options. In addition, the classification performance is reduced when using datasets with different type of stimulus, but it is similar when using datasets with different the number of symbols. These results have a special connotation for the design of systems, in which it is intended to elicit higher levels of evoked potentials and, at the same time, optimize training time.

Impact of Stimulus Features on the Performance of a Gaze-Independent Brain-Computer Interface Based on Covert Spatial Attention Shifts

Regaining communication abilities in patients who are unable to speak or move is one of the main goals in decoding brain waves for brain-computer interface (BCI) control. Many BCI approaches designed for communication rely on attention to visual stimuli, commonly applying an oddball paradigm, and require both eye movements and adequate visual acuity. These abilities may, however, be absent in patients who depend on BCI communication. We have therefore developed a response-based communication BCI, which is independent of gaze shifts but utilizes covert shifts of attention to the left or right visual field. We recorded the electroencephalogram (EEG) from 29 channels and coregistered the vertical and horizontal electrooculogram. Data-driven decoding of small attention-based differences between the hemispheres, also known as N2pc, was performed using 14 posterior channels, which are expected to reflect correlates of visual spatial attention. Eighteen healthy participants responded to 120 statements by covertly directing attention to one of two colored symbols (green and red crosses for "yes" and "no," respectively), presented in the user's left and right visual field, respectively, while maintaining central gaze fixation. On average across participants, 88.5% (std: 7.8%) of responses were correctly decoded online. In order to investigate the potential influence of stimulus features on accuracy, we presented the symbols with different visual angles, by altering symbol size and eccentricity. The offline analysis revealed that stimulus features have a minimal impact on the controllability of the BCI. Hence, we show with our novel approach that spatial attention to a colored symbol is a robust method with which to control a BCI, which has the potential to support severely paralyzed people with impaired eye movements and low visual acuity in communicating with their environment.

Evaluation of a P300-Based Brain-Machine Interface for a Robotic Hand-Orthosis Control

This work presents the design, implementation, and evaluation of a P300-based brain-machine interface (BMI) developed to control a robotic hand-orthosis. The purpose of this system is to assist patients with amyotrophic lateral sclerosis (ALS) who cannot open and close their hands by themselves. The user of this interface can select one of six targets, which represent the flexion-extension of one finger independently or the movement of the five fingers simultaneously. We tested offline and online our BMI on eighteen healthy subjects (HS) and eight ALS patients. In the offline test, we used the calibration data of each participant recorded in the experimental sessions to estimate the accuracy of the BMI to classify correctly single epochs as target or non-target trials. On average, the system accuracy was 78.7% for target epochs and 85.7% for non-target trials. Additionally, we observed significant P300 responses in the calibration recordings of all the participants, including the ALS patients. For the BMI online test, each subject performed from 6 to 36 attempts of target selections using the interface. In this case, around 46% of the participants obtained 100% of accuracy, and the average online accuracy was 89.83%. The maximum information transfer rate (ITR) observed in the experiments was 52.83 bit/min, whereas that the average ITR was 18.13 bit/min. The contributions of this work are the following. First, we report the development and evaluation of a mind-controlled robotic hand-orthosis for patients with ALS. To our knowledge, this BMI is one of the first P300-based assistive robotic devices with multiple targets evaluated on people with ALS. Second, we provide a database with calibration data and online EEG recordings obtained in the evaluation of our BMI. This data is useful to develop and compare other BMI systems and test the processing pipelines of similar applications.

Using SSVEP-BCI to Continuous Control a Quadcopter with 4-DOF Motions

Brain-computer interfaces (BCIs) allow for translating electroencephalogram (EEG) into control commands, e.g., to control a quadcopter. This study, we developed a practical BCI based on steady-state visually evoked potential (SSVEP) for continuous control of a quadcopter from the first-person perspective. Users watched with the video stream from a camera on the quadcopter. An innovative user interface was developed by embedding 12 SSVEP flickers into the video stream, which corresponded to the flight commands of 'take-off,' 'land,' 'hover,' 'keep-going,' 'clockwise,' 'counter-clockwise' and rectilinear motions in six directions, respectively. The command was updated every 400ms by decoding the collected EEG data using a combined classification algorithm based on task-related component analysis (TRCA) and linear discriminant analysis (LDA). The quadcopter flew in the 3-D space according to the control vector that was determined by the latest four commands. Three novices participated in this study. They were asked to control the quadcopter by either the brain or hands to fly through a circle and land on the target zone. As a result, the time consumption ratio of brain-control to hand-control was as low as 1.34, which means the BCI performance was close to hands. The information transfer rate reached a peak of 401.79 bits/min in the simulated online experiment. These results demonstrate the proposed SSVEP-BCI system is efficient for controlling the quadcopter.