Optimal spatial filtering of single trial EEG during imagined hand movement

Motor imagery electroencephalogram classification algorithm based on joint features in the spatial and frequency domains and instance transfer

Introduction

Motor imagery electroencephalography (MI-EEG) has significant application value in the field of rehabilitation, and is a research hotspot in the brain-computer interface (BCI) field. Due to the small training sample size of MI-EEG of a single subject and the large individual differences among different subjects, existing classification models have low accuracy and poor generalization ability in MI classification tasks.

Methods

To solve this problem, this paper proposes a electroencephalography (EEG) joint feature classification algorithm based on instance transfer and ensemble learning. Firstly, the source domain and target domain data are preprocessed, and then common space mode (CSP) and power spectral density (PSD) are used to extract spatial and frequency domain features respectively, which are combined into EEG joint features. Finally, an ensemble learning algorithm based on kernel mean matching (KMM) and transfer learning adaptive boosting (TrAdaBoost) is used to classify MI-EEG.

Results

To validate the effectiveness of the algorithm, this paper compared and analyzed different algorithms on the BCI Competition IV Dataset 2a, and further verified the stability and effectiveness of the algorithm on the BCI Competition IV Dataset 2b. The experimental results show that the algorithm has an average accuracy of 91.5% and 83.7% on Dataset 2a and Dataset 2b, respectively, which is significantly better than other algorithms.

Discussion

The statement explains that the algorithm fully exploits EEG signals and enriches EEG features, improves the recognition of the MI signals, and provides a new approach to solving the above problem.

Transformed common spatial pattern for motor imagery-based brain-computer interfaces

Objective

The motor imagery (MI)-based brain-computer interface (BCI) is one of the most popular BCI paradigms. Common spatial pattern (CSP) is an effective algorithm for decoding MI-related electroencephalogram (EEG) patterns. However, it highly depends on the selection of EEG frequency bands. To address this problem, previous researchers often used a filter bank to decompose EEG signals into multiple frequency bands before applying the traditional CSP.

Approach

This study proposed a novel method, i.e., transformed common spatial pattern (tCSP), to extract the discriminant EEG features from multiple frequency bands after but not before CSP. To verify its effectiveness, we tested tCSP on a dataset collected by our team and a public dataset from BCI competition III. We also performed an online evaluation of the proposed method.

Main results

As a result, for the dataset collected by our team, the classification accuracy of tCSP was significantly higher than CSP by about 8% and filter bank CSP (FBCSP) by about 4.5%. The combination of tCSP and CSP further improved the system performance with an average accuracy of 84.77% and a peak accuracy of 100%. For dataset IVa in BCI competition III, the combination method got an average accuracy of 94.55%, which performed best among all the presented CSP-based methods. In the online evaluation, tCSP and the combination method achieved an average accuracy of 80.00 and 84.00%, respectively.

Significance

The results demonstrate that the frequency band selection after CSP is better than before for MI-based BCIs. This study provides a promising approach for decoding MI EEG patterns, which is significant for the development of BCIs.

Recognition of single upper limb motor imagery tasks from EEG using multi-branch fusion convolutional neural network

Motor imagery-based brain-computer interfaces (MI-BCI) have important application values in the field of neurorehabilitation and robot control. At present, MI-BCI mostly use bilateral upper limb motor tasks, but there are relatively few studies on single upper limb MI tasks. In this work, we conducted studies on the recognition of motor imagery EEG signals of the right upper limb and proposed a multi-branch fusion convolutional neural network (MF-CNN) for learning the features of the raw EEG signals as well as the two-dimensional time-frequency maps at the same time. The dataset used in this study contained three types of motor imagery tasks: extending the arm, rotating the wrist, and grasping the object, 25 subjects were included. In the binary classification experiment between the grasping object and the arm-extending tasks, MF-CNN achieved an average classification accuracy of 78.52% and kappa value of 0.57. When all three tasks were used for classification, the accuracy and kappa value were 57.06% and 0.36, respectively. The comparison results showed that the classification performance of MF-CNN is higher than that of single CNN branch algorithms in both binary-class and three-class classification. In conclusion, MF-CNN makes full use of the time-domain and frequency-domain features of EEG, can improve the decoding accuracy of single limb motor imagery tasks, and it contributes to the application of MI-BCI in motor function rehabilitation training after stroke.

Feature optimization based on improved novel global harmony search algorithm for motor imagery electroencephalogram classification

Background

Effectively decoding electroencephalogram (EEG) pattern for specific mental tasks is a crucial topic in the development of brain-computer interface (BCI). Extracting common spatial pattern (CSP) features from motor imagery EEG signals is often highly dependent on the selection of frequency band and time interval. Therefore, optimizing frequency band and time interval would contribute to effective feature extraction and accurate EEG decoding.

Objective

This study proposes an approach based on an improved novel global harmony search (INGHS) to optimize frequency-time parameters for effective CSP feature extraction.

Methods

The INGHS algorithm is applied to find the optimal frequency band and temporal interval. The linear discriminant analysis and support vector machine are used for EEG pattern decoding. Extensive experimental studies are conducted on three EEG datasets to assess the effectiveness of our proposed method.

Results

The average test accuracy obtained by the time-frequency parameters selected by the proposed INGHS method is slightly better than artificial bee colony (ABC) and particle swarm optimization (PSO) algorithms. Furthermore, the INGHS algorithm is superior to PSO and ABC in running time.

Conclusion

These superior experimental results demonstrate that the optimal frequency band and time interval selected by the INGHS algorithm could significantly improve the decoding accuracy compared with the traditional CSP method. This method has a potential to improve the performance of MI-based BCI systems.

A novel heterogeneous transfer learning method based on data stitching for the sequential coding brain computer interface

OBJECTIVE

For the brain computer interface (BCI), it is necessary to collect enough electroencephalography (EEG) signals to train the classification model. When the operation dimension of BCI is large, it will bring great burden to data acquisition. Fortunately, this problem can be solved by our proposed transfer learning method.

METHOD

For the sequential coding experimental paradigm, the multi-band data stitching with label alignment and tangent space mapping (MDSLATSM) algorithm is proposed as a novel heterogeneous transfer learning method. After filtering by multi-band filtering, the artificial signals can be obtained by data stitching from the source domain, which build a bridge between the source domain and target domain. To make the distribution of two domains closer, their covariance matrices are aligned by label alignment. After mapping to the tangent space, the features are extracted from the Riemannian manifold. Finally, the classification results are obtained with feature selection and classification.

RESULTS

Our data set includes the EEG signals from 16 subjects. For the heterogeneous transfer learning of cross-label, the average classification accuracy is 78.28%. MDSLATSM is also tested for cross-subject, and the average classification accuracy is 64.01%, which is better than existing methods.

SIGNIFICANCE

Combining multi-band filtering, data stitching, label alignment and tangent space mapping, a novel heterogeneous transfer learning method can be achieved with superior performance, which promotes the practical application of the BCI systems.

Segment alignment based cross-subject motor imagery classification under fading data

Motor imagery (MI) aims to use brain imagination without actual body activities to support motor learning, and machine learning algorithms such as common spatial patterns (CSP) are proven effective in the analysis of MI signals. In the conventional machine learning-based approaches, there are two main difficulties in feature extraction and recognition of MI signals: high personalization and data fading. The high personalization problem is due to the multi-subject nature when collecting MI signals, and the data fading problem as a recurring issue in MI signal quality is first raised by us but is not widely discussed at present. Aiming to solve the above two mentioned problems, a cross-subject fading data classification approach with segment alignment is proposed to classify the fading data of one single target with the model trained with the normal data of multiple sources in this paper. he effectiveness of proposed method is verified via two experiments: a dataset-based experiment with the dataset from BCI Competition and a lab-based experiment designed and conducted by us. The experimental results obtained from both experiments show that the proposed method can obtain optimal classification performance effectively under different fading levels with data from different subjects.

Data augmentation for learning predictive models on EEG: a systematic comparison

Objective.The use of deep learning for electroencephalography (EEG) classification tasks has been rapidly growing in the last years, yet its application has been limited by the relatively small size of EEG datasets. Data augmentation, which consists in artificially increasing the size of the dataset during training, can be employed to alleviate this problem. While a few augmentation transformations for EEG data have been proposed in the literature, their positive impact on performance is often evaluated on a single dataset and compared to one or two competing augmentation methods. This work proposes to better validate the existing data augmentation approaches through a unified and exhaustive analysis.Approach.We compare quantitatively 13 different augmentations with two different predictive tasks, datasets and models, using three different types of experiments.Main results.We demonstrate that employing the adequate data augmentations can bring up to 45% accuracy improvements in low data regimes compared to the same model trained without any augmentation. Our experiments also show that there is no single best augmentation strategy, as the good augmentations differ on each task.Significance.Our results highlight the best data augmentations to consider for sleep stage classification and motor imagery brain-computer interfaces. More broadly, it demonstrates that EEG classification tasks benefit from adequate data augmentation.

EEG-based mental fatigue detection using linear prediction cepstral coefficients and Riemann spatial covariance matrix

Objective. Establishing a mental fatigue monitoring system is of great importance as for severe fatigue may cause unimaginable consequences. Electroencephalogram (EEG) is often utilized for mental fatigue detection because of its high temporal resolution and ease of use. However, many EEG-based approaches for detecting mental fatigue only take into account the feature extraction of a single domain and do not fully exploit the information that EEG may offer.Approach. In our work, we propose a new algorithm for mental fatigue detection based on multi-domain feature extraction and fusion. EEG components representing fatigue are closely related in the past and present because fatigue is a dynamic and gradual process. Accordingly, the idea of linear prediction is used to fit the current value with a set of sample values in the past to calculate the linear prediction cepstral coefficients (LPCCs) as the time domain feature. Moreover, in order to better capture fatigue-related spatial domain information, the spatial covariance matrix of the original EEG signal is projected into the Riemannian tangent space using the Riemannian geometric method. Then multi-domain features are fused to obtain comprehensive spatio-temporal information.Main results. Experimental results prove the suggested algorithm outperforms existing state-of-the-art methods, achieving an average accuracy of 87.10% classification on the public dataset SEED-VIG (three categories) and 97.40% classification accuracy (two categories) on the dataset made by self-designed experiments.Significance. These findings show that our proposed strategy perform more effectively for mental fatigue detection based on EEG.

Channel Reduction for an EEG-Based Authentication System While Performing Motor Movements

Commercial use of biometric authentication is becoming increasingly popular, which has sparked the development of EEG-based authentication. To stimulate the brain and capture characteristic brain signals, these systems generally require the user to perform specific activities such as deeply concentrating on an image, mental activity, visual counting, etc. This study investigates whether effective authentication would be feasible for users tasked with a minimal daily activity such as lifting a tiny object. With this novel protocol, the minimum number of EEG electrodes (channels) with the highest performance (ranked) was identified to improve user comfort and acceptance over traditional 32-64 electrode-based EEG systems while also reducing the load of real-time data processing. For this proof of concept, a public dataset was employed, which contains 32 channels of EEG data from 12 participants performing a motor task without intent for authentication. The data was filtered into five frequency bands, and 12 different features were extracted to train a random forest-based machine learning model. All channels were ranked according to Gini Impurity. It was found that only 14 channels are required to perform authentication when EEG data is filtered into the Gamma sub-band within a 1% accuracy of using 32-channels. This analysis will allow (a) the design of a custom headset with 14 electrodes clustered over the frontal and occipital lobe of the brain, (b) a reduction in data collection difficulty while performing authentication, (c) minimizing dataset size to allow real-time authentication while maintaining reasonable performance, and (d) an API for use in ranking authentication performance in different headsets and tasks.

A Survey on Physiological Signal-Based Emotion Recognition

Physiological signals are the most reliable form of signals for emotion recognition, as they cannot be controlled deliberately by the subject. Existing review papers on emotion recognition based on physiological signals surveyed only the regular steps involved in the workflow of emotion recognition such as pre-processing, feature extraction, and classification. While these are important steps, such steps are required for any signal processing application. Emotion recognition poses its own set of challenges that are very important to address for a robust system. Thus, to bridge the gap in the existing literature, in this paper, we review the effect of inter-subject data variance on emotion recognition, important data annotation techniques for emotion recognition and their comparison, data pre-processing techniques for each physiological signal, data splitting techniques for improving the generalization of emotion recognition models and different multimodal fusion techniques and their comparison. Finally, we discuss key challenges and future directions in this field.

Subject adaptation convolutional neural network for EEG-based motor imagery classification

Objective.Deep transfer learning has been widely used to address the nonstationarity of electroencephalogram (EEG) data during motor imagery (MI) classification. However, previous deep learning approaches suffer from limited classification accuracy because the temporal and spatial features cannot be effectively extracted.Approach.Here, we propose a novel end-to-end deep subject adaptation convolutional neural network (SACNN) to handle the problem of EEG-based MI classification. Our proposed model jointly optimizes three modules, i.e. a feature extractor, a classifier, and a subject adapter. Specifically, the feature extractor simultaneously extracts the temporal and spatial features from the raw EEG data using a parallel multiscale convolution network. In addition, we design a subject adapter to reduce the feature distribution shift between the source and target subjects by using the maximum mean discrepancy. By minimizing the classification loss and the distribution discrepancy, the model is able to extract the temporal-spatial features to the prediction of a new subject.Main results.Extensive experiments are carried out on three EEG-based MI datasets, i.e. brain-computer interface (BCI) competition IV dataset IIb, BCI competition III dataset IVa, and BCI competition IV dataset I, and the average accuracy reaches to 86.42%, 81.71% and 79.35% on the three datasets respectively. Furthermore, the statistical analysis also indicates the significant performance improvement of SACNN.Significance.This paper reveals the importance of the temporal-spatial features on EEG-based MI classification task. Our proposed SACNN model can make fully use of the temporal-spatial information to achieve the purpose.

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.

Learning Optimal Time-Frequency-Spatial Features by the CiSSA-CSP Method for Motor Imagery EEG Classification

The common spatial pattern (CSP) is a popular method in feature extraction for motor imagery (MI) electroencephalogram (EEG) classification in brain-computer interface (BCI) systems. However, combining temporal and spectral information in the CSP-based spatial features is still a challenging issue, which greatly affects the performance of MI-based BCI systems. Here, we propose a novel circulant singular spectrum analysis embedded CSP (CiSSA-CSP) method for learning the optimal time-frequency-spatial features to improve the MI classification accuracy. Specifically, raw EEG data are first segmented into multiple time segments and spectrum-specific sub-bands are further derived by CiSSA from each time segment in a set of non-overlapping filter bands. CSP features extracted from all time-frequency segments contain more sufficient time-frequency-spatial information. An experimental study was implemented on the publicly available EEG dataset (BCI Competition III dataset IVa) and a self-collected experimental EEG dataset to validate the effectiveness of the CiSSA-CSP method. Experimental results demonstrate that discriminative and robust features are extracted effectively. Compared with several state-of-the-art methods, the proposed method exhibited optimal accuracies of 96.6% and 95.2% on the public and experimental datasets, respectively, which confirms that it is a promising method for improving the performance of MI-based BCIs.

Use of common spatial patterns for early detection of Parkinson's disease

One of the most common diseases that affects human brain is Parkinson's disease. Detection of Parkinson's disease (PD) poses a serious challenge. Robust methods for feature extraction allowing separation between the electroencephalograms (EEG) of healthy subjects and PD patients are required. We used the EEG records of healthy subjects and PD patients which were subject to auditory tasks. We used the common spatial patterns (CSP) and Laplacian mask as methods to allow robust selection and extraction of features. We used the derived CSP whitening matrix to determine those channels that are the most promising in the terms of differentiating between EEGs of healthy controls and of PD patients. Using the selection of features calculated using the CSP we managed to obtain the classification accuracy of 85% when classifying EEG records belonging to groups of controls or PD patients. Using the features calculated using the Laplacian operator we obtained the classification accuracy of 90%. Diagnosing the PD in early stages using EEG is possible. The CSP proved to be a promising technique to detect informative channels and to separate between the groups. Use of the combination of features calculated using the Laplacian offers good separability between the two groups.

Mental Strategies in a P300-BCI: Visuomotor Transformation Is an Option

Currently, P300-BCIs are mostly used for spelling tasks, where the number of commands is equal to the number of stimuli that evoke event-related potentials (ERPs). Increasing this number slows down the BCI operation because each stimulus has to be presented several times for better classification. Furthermore, P300 spellers typically do not utilize potentially useful imagery-based approaches, such as the motor imagery successfully practiced in motor rehabilitation. Here, we tested a P300-BCI with a motor-imagery component. In this BCI, the number of commands was increased by adding mental strategies instead of increasing the number of targets. Our BCI had only two stimuli and four commands. The subjects either counted target appearances mentally or imagined hand movements toward the targets. In this design, the motor-imagery paradigm enacted a visuomotor transformation known to engage cortical and subcortical networks participating in motor control. The operation of these networks suffers in neurological conditions such as stroke, so we view this BCI as a potential tool for the rehabilitation of patients. As an initial step toward the development of this clinical method, sixteen healthy participants were tested. Consistent with our expectation that mental strategies would result in distinct EEG activities, ERPs were different depending on whether subjects counted stimuli or imagined movements. These differences were especially clear in the late ERP components localized in the frontal and centro-parietal regions. We conclude that (1) the P300 paradigm is suitable for enacting visuomotor transformations and (2) P300-based BCIs with multiple mental strategies could be used in applications where the number of possible outputs needs to be increased while keeping the number of targets constant. As such, our approach adds to both the development of versatile BCIs and clinical approaches to rehabilitation.

Multi-Kernel Temporal and Spatial Convolution for EEG-Based Emotion Classification

Deep learning using an end-to-end convolutional neural network (ConvNet) has been applied to several electroencephalography (EEG)-based brain-computer interface tasks to extract feature maps and classify the target output. However, the EEG analysis remains challenging since it requires consideration of various architectural design components that influence the representational ability of extracted features. This study proposes an EEG-based emotion classification model called the multi-kernel temporal and spatial convolution network (MultiT-S ConvNet). The multi-scale kernel is used in the model to learn various time resolutions, and separable convolutions are applied to find related spatial patterns. In addition, we enhanced both the temporal and spatial filters with a lightweight gating mechanism. To validate the performance and classification accuracy of MultiT-S ConvNet, we conduct subject-dependent and subject-independent experiments on EEG-based emotion datasets: DEAP and SEED. Compared with existing methods, MultiT-S ConvNet outperforms with higher accuracy results and a few trainable parameters. Moreover, the proposed multi-scale module in temporal filtering enables extracting a wide range of EEG representations, covering short- to long-wavelength components. This module could be further implemented in any model of EEG-based convolution networks, and its ability potentially improves the model's learning capacity.

EEG channel selection based on sequential backward floating search for motor imagery classification

Brain-computer interfaces (BCIs) based on motor imagery (MI) utilizing multi-channel electroencephalogram (EEG) data are commonly used to improve motor function of people with motor disabilities. EEG channel selection can enhance MI classification accuracy by selecting informative channels, accordingly reducing redundant information. The sequential backward floating search (SBFS) approach has been considered as one of the best feature selection methods. In this paper, SBFS is first implemented to select the optimal EEG channels in MI-BCI. Further, to reduce the time complexity of SBFS, the modified SBFS is proposed and applied to left and right hand MI tasks. In the modified SBFS, based on the map of EEG channels at the scalp, the symmetrical channels are selected as channel pairs and acceleration is thus realized by removing or adding multiple channels in each iteration. Extensive experiments were conducted on four public BCI datasets. Experimental results show that the SBFS achieves significantly higher classification accuracy (p < 0.001) than using all channels and conventional MI channels (i.e., C3, C4, and Cz). Moreover, the proposed method outperforms the state-of-the-art selection methods.

Supervised and Semisupervised Manifold Embedded Knowledge Transfer in Motor Imagery-Based BCI

A long calibration procedure limits the use in practice for a motor imagery (MI)-based brain-computer interface (BCI) system. To tackle this problem, we consider supervised and semisupervised transfer learning. However, it is a challenge for them to cope with high intersession/subject variability in the MI electroencephalographic (EEG) signals. Based on the framework of unsupervised manifold embedded knowledge transfer (MEKT), we propose a supervised MEKT algorithm (sMEKT) and a semisupervised MEKT algorithm (ssMEKT), respectively. sMEKT only has limited labelled samples from a target subject and abundant labelled samples from multiple source subjects. Compared to sMEKT, ssMEKT adds comparably more unlabelled samples from the target subject. After performing Riemannian alignment (RA) and tangent space mapping (TSM), both sMEKT and ssMEKT execute domain adaptation to shorten the differences among subjects. During domain adaptation, to make use of the available samples, two algorithms preserve the source domain discriminability, and ssMEKT preserves the geometric structure embedded in the labelled and unlabelled target domains. Moreover, to obtain a subject-specific classifier, sMEKT minimizes the joint probability distribution shift between the labelled target and source domains, whereas ssMEKT performs the joint probability distribution shift minimization between the unlabelled target domain and all labelled domains. Experimental results on two publicly available MI datasets demonstrate that our algorithms outperform the six competing algorithms, where the sizes of labelled and unlabelled target domains are variable. Especially for the target subjects with 10 labelled samples and 270/190 unlabelled samples, ssMEKT shows 5.27% and 2.69% increase in average accuracy on the two abovementioned datasets compared to the previous best semisupervised transfer learning algorithm (RA-regularized common spatial patterns-weighted adaptation regularization, RA-RCSP-wAR), respectively. Therefore, our algorithms can effectively reduce the need of labelled samples for the target subject, which is of importance for the MI-based BCI application.

Exploring the ability of stroke survivors in using the contralesional hemisphere to control a brain-computer interface

Brain-computer interfaces (BCIs) have recently been shown to be clinically effective as a novel method of stroke rehabilitation. In many BCI-based studies, the activation of the ipsilesional hemisphere was considered a key factor required for motor recovery after stroke. However, emerging evidence suggests that the contralesional hemisphere also plays a role in motor function rehabilitation. The objective of this study is to investigate the effectiveness of the BCI in detecting motor imagery of the affected hand from contralesional hemisphere. We analyzed a large EEG dataset from 136 stroke patients who performed motor imagery of their stroke-impaired hand. BCI features were extracted from channels covering either the ipsilesional, contralesional or bilateral hemisphere, and the offline BCI accuracy was computed using 10 [Formula: see text] 10-fold cross-validations. Our results showed that most stroke patients can operate the BCI using either their contralesional or ipsilesional hemisphere. Those with the ipsilesional BCI accuracy of less than 60% had significantly higher motor impairments than those with the ipsilesional BCI accuracy above 80%. Interestingly, those with the ipsilesional BCI accuracy of less than 60% achieved a significantly higher contralesional BCI accuracy, whereas those with the ipsilesional BCI accuracy more than 80% had significantly poorer contralesional BCI accuracy. This study suggests that contralesional BCI may be a useful approach for those with a high motor impairment who cannot accurately generate signals from ipsilesional hemisphere to effectively operate BCI.

Research on Movement Intentions of Human's Left and Right Legs Based on EEG Signals

Active rehabilitation training method can help stroke patients recover better and faster. However, the lower limb rehabilitation robot based on electroencephalogram (EEG) has low recognition accuracy now. A classification method based on EEG signals of motor imagery is proposed to enable patients to accurately control their left and right legs. Firstly, aiming at the unstable characteristics of EEG signals, an experimental protocl of motor imagery was constructed based on multi-joint motion coupling of left and right legs. The signals with time-frequency analysis and ERD/S analysis have proved the reliability and validity of the collected EEG signals. Then, the EEG signals generated by the protocol were preprocessed and Common Space Pattern (CSP) was used to extract their features. Support Vector Machine (SVM) and Linear Discriminant Analysis (LDA) are adapted and their accuracy of classification results are compared. Finally, on the basis of the proposed classifier with excellent performance, the classifier is used in the active control strategy of the lower limb rehabilitation robot, and the experiment verified that the average accuracy of two volunteers in controlling the lower limb rehabilitation robot reached 95.1%. This research provides a good theoretical basis for the realization and application of brain-computer interface in rehabilitation training.

A large EEG dataset for studying cross-session variability in motor imagery brain-computer interface

In building a practical and robust brain-computer interface (BCI), the classification of motor imagery (MI) from electroencephalography (EEG) across multiple days is a long-standing challenge due to the large variability of the EEG signals. We collected a large dataset of MI from 5 different days with 25 subjects, the first open-access dataset to address BCI issues across 5 different days with a large number of subjects. The dataset includes 5 session data from 5 different days (2–3 days apart) for each subject. Each session contains 100 trials of left-hand and right-hand MI. In this report, we provide the benchmarking classification accuracy for three conditions, namely, within-session classification (WS), cross-session classification (CS), and cross-session adaptation (CSA), with subject-specific models. WS achieves an average classification accuracy of up to 68.8%, while CS degrades the accuracy to 53.7% due to the cross-session variability. However, by adaptation, CSA improves the accuracy to 78.9%. We anticipate this new dataset will significantly push further progress in MI BCI research in addressing the cross-session and cross-subject challenge.

Measurement(s)	Electroencephalography
Technology Type(s)	motor imagery

Deep-Learning-Based Automatic Selection of Fewest Channels for Brain-Machine Interfaces

Due to the development of convenient brain-machine interfaces (BMIs), the automatic selection of a minimum channel (electrode) set has attracted increasing interest because the decrease in the number of channels increases the efficiency of BMIs. This study proposes a deep-learning-based technique to automatically search for the minimum number of channels applicable to general BMI paradigms using a compact convolutional neural network for electroencephalography (EEG)-based BMIs. For verification, three types of BMI paradigms are assessed: 1) the typical P300 auditory oddball; 2) the new top-down steady-state visually evoked potential; and 3) the endogenous motor imagery. We observe that the optimized minimal EEG-channel sets are automatically selected in all three cases. Their decoding accuracies using the minimal channels are statistically equivalent to (or even higher than) those based on all channels. The brain areas of the selected channel set are neurophysiologically interpretable for all of these cognitive task paradigms. This study shows that the minimal EEG channel set can be automatically selected, irrespective of the types of BMI paradigms or EEG input features using a deep-learning approach, which also contributes to their portability.

An Effective Fusing Approach by Combining Connectivity Network Pattern and Temporal-Spatial Analysis for EEG-Based BCI Rehabilitation

Motor-modality-based brain computer interface (BCI) could promote the neural rehabilitation for stroke patients. Temporal-spatial analysis was commonly used for pattern recognition in this task. This paper introduced a novel connectivity network analysis for EEG-based feature selection. The network features of connectivity pattern not only captured the spatial activities responding to motor task, but also mined the interactive pattern among these cerebral regions. Furthermore, the effective combination between temporal-spatial analysis and network analysis was evaluated for improving the performance of BCI classification (81.7%). And the results demonstrated that it could raise the classification accuracies for most of patients (6 of 7 patients). This proposed method was meaningful for developing the effective BCI training program for stroke rehabilitation.

On the Deep Learning Models for EEG-Based Brain-Computer Interface Using Motor Imagery

Motor imagery (MI) based brain-computer interface (BCI) is an important BCI paradigm which requires powerful classifiers. Recent development of deep learning technology has prompted considerable interest in using deep learning for classification and resulted in multiple models. Finding the best performing models among them would be beneficial for designing better BCI systems and classifiers going forward. However, it is difficult to directly compare performance of various models through the original publications, since the datasets used to test the models are different from each other, too small, or even not publicly available. In this work, we selected five MI-EEG deep classification models proposed recently: EEGNet, Shallow & Deep ConvNet, MB3D and ParaAtt, and tested them on two large, publicly available, databases with 42 and 62 human subjects. Our results show that the models performed similarly on one dataset while EEGNet performed the best on the second with a relatively small training cost using the parameters that we evaluated.

Performance Variation of a Somatosensory BCI Based on Imagined Sensation: A Large Population Study

A proportion of users cannot achieve adequate brain-computer interface (BCI) control. The diversity of BCI modalities provides a way to solve this emerging issue. Here, we investigate the accuracy of a somatosensory BCI based on sensory imagery (SI). During the SI tasks, subjects were instructed to imagine a tactile sensation and to maintain the attention on the corresponding hand, as if there was tactile stimulus on the skin of the wrist. The performance across 106 healthy subjects in left- and right-hand SI discrimination was 78.9±13.2%. In 70.7% of the subjects the performance was above 70%. The SI task induced a contralateral cortical activation, and high-density EEG source localization showed that the real tactile stimulation and imagined tactile stimulation shared similar cortical activations within the somatosensory cortex. The somatosensory BCI based on SI provides a new signal modality for independent BCI development. Moreover, a combination of SI and other BCI modalities, such as motor imagery, may provide new avenues for further improving BCI usage and applicability, especially in those subjects unable to attain adequate BCI control with conventional BCI modalities.

A Fused Multidimensional EEG Classification Method Based on an Extreme Tree Feature Selection

When a brain-computer interface (BCI) is designed, high classification accuracy is difficult to obtain for motor imagery (MI) electroencephalogram (EEG) signals in view of their relatively low signal-to-noise ratio. In this paper, a fused multidimensional classification method based on extreme tree feature selection (FMCM-ETFS) is proposed for discerning motor imagery EEG tasks. First, the EEG signal was filtered by a Butterworth filter for preprocessing. Second, C3, C4, and CZ channels were selected to extract time-frequency domain and spatial domain features using autoregressive (AR), common spatial pattern (CSP), and discrete wavelet transform (DWT). The extracted features were fused for a further feature elimination. Then, the features were selected using three feature selection methods: recursive feature elimination (RFE), principal component analysis method (PCA), and extreme trees (ET). The selected feature vectors were classified using support vector machines (SVM). Finally, a total of twelve subjects’ EEG data from Inner Mongolia University of Technology (IMUT data), the 2nd BCI competition in 2003, and the 4th BCI competition in 2008 were employed to show the effectiveness of this proposed FMCM-ETFS method. The results show that the classification accuracy using the multidimensional fused feature extraction (AR + CSP + DWT) is 3%–20% higher than those using the aforementioned three single feature extractions (AR, CSP, and DWT). Extreme trees (ET), which is a sort of tree-based model method, outperforms RFE and PCA by 1%–9% in term of classification accuracies, when these three methods were applied to the procedure of feature extraction, respectively.

Miniaturization for wearable EEG systems: recording hardware and data processing

As more people desire at-home diagnosis and treatment for their health improvement, healthcare devices have become more wearable, comfortable, and easy to use. In that sense, the miniaturization of electroencephalography (EEG) systems is a major challenge for developing daily-life healthcare devices. Recently, because of the intertwined relationship between EEG recording and processing, co-research of EEG recording hardware and data processing has been emphasized for whole-in-one miniaturized EEG systems. This paper introduces miniaturization techniques in analog-front-end hardware and processing algorithms for such EEG systems. To miniaturize EEG recording hardware, various types of compact electrodes and mm-sized integrated circuits (IC) techniques including artifact rejection are studied to record accurate EEG signals in a much smaller manner. Active electrode and in-ear EEG technologies are also researched to make small-form-factor EEG measurement structures. Furthermore, miniaturization techniques for EEG processing are discussed including channel selection techniques that reduce the number of required electrode channel and hardware implementation of processing algorithms that simplify the EEG processing stage.

Decoding locomotion speed and slope from local field potentials of rat motor cortex

BACKGROUND AND OBJECTIVE

Local Field Potentials (LFPs) recorded from the primary motor cortex (M1) have been shown to be very informative for decoding movement parameters, and these signals can be used to decode forelimb kinematic and kinetic parameters accurately. Although locomotion is one of the most basic and important motor abilities of humans and animals, the potential of LFPs in decoding abstract hindlimb locomotor parameters has not been investigated. This study investigates the feasibility of decoding speed and slope of locomotion, as two important abstract parameters of walking, using the LFP signals.

METHODS

Rats were trained to walk smoothly on a treadmill with different speeds and slopes. The brain signals were recorded using the microwire arrays chronically implanted in the hindlimb area of M1 while rats walked on the treadmill. LFP channels were spatially filtered using optimal common spatial patterns to increase the discriminability of speeds and slopes of locomotion. Logarithmic wavelet band powers were extracted as basic features, and the best features were selected using the statistical dependency criterion before classification.

RESULTS

Using 5 s LFP trials, the average classification accuracies of four different speeds and seven different slopes reached 90.8% and 86.82%, respectively. The high-frequency LFP band (250-500 Hz) was the most informative band about these parameters and contributed more than other frequency bands in the final decoder model.

CONCLUSIONS

Our results show that the LFP signals in M1 accurately decode locomotion speed and slope, which can be considered as abstract walking parameters needed for designing long-term brain-computer interfaces for hindlimb locomotion control.

Improving EEG Decoding via Clustering-Based Multitask Feature Learning

Accurate electroencephalogram (EEG) pattern decoding for specific mental tasks is one of the key steps for the development of brain-computer interface (BCI), which is quite challenging due to the considerably low signal-to-noise ratio of EEG collected at the brain scalp. Machine learning provides a promising technique to optimize EEG patterns toward better decoding accuracy. However, existing algorithms do not effectively explore the underlying data structure capturing the true EEG sample distribution and, hence, can only yield a suboptimal decoding accuracy. To uncover the intrinsic distribution structure of EEG data, we propose a clustering-based multitask feature learning algorithm for improved EEG pattern decoding. Specifically, we perform affinity propagation-based clustering to explore the subclasses (i.e., clusters) in each of the original classes and then assign each subclass a unique label based on a one-versus-all encoding strategy. With the encoded label matrix, we devise a novel multitask learning algorithm by exploiting the subclass relationship to jointly optimize the EEG pattern features from the uncovered subclasses. We then train a linear support vector machine with the optimized features for EEG pattern decoding. Extensive experimental studies are conducted on three EEG data sets to validate the effectiveness of our algorithm in comparison with other state-of-the-art approaches. The improved experimental results demonstrate the outstanding superiority of our algorithm, suggesting its prominent performance for EEG pattern decoding in BCI applications.

Reducing the Calibration Time in Somatosensory BCI by Using Tactile ERD

OBJECTIVE

We propose a tactile-induced-oscillation approach to reduce the calibration time in somatosensory brain-computer interfaces (BCI).

METHODS

Based on the similarity between tactile induced event-related desynchronization (ERD) and imagined sensation induced ERD activation, we extensively evaluated BCI performance when using a conventional and a novel calibration strategy. In the conventional calibration, the tactile imagined data was used, while in the sensory calibration model sensory stimulation data was used. Subjects were required to sense the tactile stimulus when real tactile was applied to the left or right wrist and were required to perform imagined sensation tasks in the somatosensory BCI paradigm.

RESULTS

The sensory calibration led to a significantly better performance than the conventional calibration when tested on the same imagined sensation dataset ( [Formula: see text]=10.89, P=0.0038), with an average 5.1% improvement in accuracy. Moreover, the sensory calibration was 39.3% faster in reaching a performance level of above 70% accuracy.

CONCLUSION

The proposed approach of using tactile ERD from the sensory cortex provides an effective way of reducing the calibration time in a somatosensory BCI system.

SIGNIFICANCE

The tactile stimulation would be specifically useful before BCI usage, avoiding excessive fatigue when the mental task is difficult to perform. The tactile ERD approach may find BCI applications for patients or users with preserved afferent pathways.

Detection of Stroke-Induced Visual Neglect and Target Response Prediction Using Augmented Reality and Electroencephalography

We aim to build a system incorporating electroencephalography (EEG) and augmented reality (AR) that is capable of identifying the presence of visual spatial neglect (SN) and mapping the estimated neglected visual field. An EEG-based brain-computer interface (BCI) was used to identify those spatiospectral features that best detect participants with SN among stroke survivors using their EEG responses to ipsilesional and contralesional visual stimuli. Frontal-central delta and alpha, frontal-parietal theta, Fp1 beta, and left frontal gamma were found to be important features for neglect detection. Additionally, temporal analysis of the responses shows that the proposed model is accurate in detecting potentially neglected targets. These targets were predicted using common spatial patterns as the feature extraction algorithm and regularized discriminant analysis combined with kernel density estimation for classification. With our preliminary results, our system shows promise for reliably detecting the presence of SN and predicting visual target responses in stroke patients with SN.

Evaluation of Multi-layer Perceptron Neural Networks in Predicting Ankle Dorsiflexion in Healthy Adults using Movement-related Cortical Potentials for BCI-Neurofeedback Applications

Brain computer interface (BCI) systems were initially developed to replace lost function; however, they are being increasingly utilized in rehabilitation to restore motor functioning after brain injury. In such BCI-mediated neurofeedback training (BCI-NFT), the brain-state associated with movement attempt or intention is used to activate an external device which assists the movement while providing sensory feedback to enhance neuroplasticity. A critical element in the success of BCI-NFT is accurate timing of the feedback within the active period of the brain state. The overarching goal of this work was to develop a reliable deep learning model that can predict motion before its onset, and thereby deliver the sensory stimuli in a timely manner for BCI-NFT applications. To this end, the main objective of the current study was to design and evaluate a Multi-layer Perceptron Neural Network (MLP-NN). Movement-related cortical potentials (MRCP) during planning and execution of ankle dorsiflexion was used to train the model to classify dorsiflexion planning vs. rest. The accuracy and reliability of the model was evaluated offline using data from eight healthy individuals (age: 26.3 ± 7.6 years). First, we evaluated three different epoching strategies for defining our 2 classes, to identify the one which best discriminated rest from dorsiflexion. The best model accuracy for predicting ankle dorsiflexion from EEG before movement execution was 84.7%. Second, the effect of various spatial filters on the model accuracy was evaluated, demonstrating that the spatial filtering had minimal effect on model accuracy and reliability.

Prototype-based Domain Generalization Framework for Subject-Independent Brain-Computer Interfaces

Brain-computer interface (BCI) is challenging to use in practice due to the inter/intra-subject variability of electroencephalography (EEG). The BCI system, in general, necessitates a calibration technique to obtain subject/session-specific data in order to tune the model each time the system is utilized. This issue is acknowledged as a key hindrance to BCI, and a new strategy based on domain generalization has recently evolved to address it. In light of this, we've concentrated on developing an EEG classification framework that can be applied directly to data from unknown domains (i.e. subjects), using only data acquired from separate subjects previously. For this purpose, in this paper, we proposed a framework that employs the open-set recognition technique as an auxiliary task to learn subject-specific style features from the source dataset while helping the shared feature extractor with mapping the features of the unseen target dataset as a new unseen domain. Our aim is to impose cross-instance style in-variance in the same domain and reduce the open space risk on the potential unseen subject in order to improve the generalization ability of the shared feature extractor. Our experiments showed that using the domain information as an auxiliary network increases the generalization performance. Clinical relevance-This study suggests a strategy to improve the performance of the subject-independent BCI systems. Our framework can help to reduce the need for further calibration and can be utilized for a range of mental state monitoring tasks (e.g. neurofeedback, identification of epileptic seizures, and sleep disorders).

Spatio-Spectral Feature Representation for Motor Imagery Classification Using Convolutional Neural Networks

Convolutional neural networks (CNNs) have recently been applied to electroencephalogram (EEG)-based brain-computer interfaces (BCIs). EEG is a noninvasive neuroimaging technique, which can be used to decode user intentions. Because the feature space of EEG data is highly dimensional and signal patterns are specific to the subject, appropriate methods for feature representation are required to enhance the decoding accuracy of the CNN model. Furthermore, neural changes exhibit high variability between sessions, subjects within a single session, and trials within a single subject, resulting in major issues during the modeling stage. In addition, there are many subject-dependent factors, such as frequency ranges, time intervals, and spatial locations at which the signal occurs, which prevent the derivation of a robust model that can achieve the parameterization of these factors for a wide range of subjects. However, previous studies did not attempt to preserve the multivariate structure and dependencies of the feature space. In this study, we propose a method to generate a spatiospectral feature representation that can preserve the multivariate information of EEG data. Specifically, 3-D feature maps were constructed by combining subject-optimized and subject-independent spectral filters and by stacking the filtered data into tensors. In addition, a layer-wise decomposition model was implemented using our 3-D-CNN framework to secure reliable classification results on a single-trial basis. The average accuracies of the proposed model were 87.15% (±7.31), 75.85% (±12.80), and 70.37% (±17.09) for the BCI competition data sets IV_2a, IV_2b, and OpenBMI data, respectively. These results are better than those obtained by state-of-the-art techniques, and the decomposition model obtained the relevance scores for neurophysiologically plausible electrode channels and frequency domains, confirming the validity of the proposed approach.

Nuclear Norm Regularized Deep Neural Network for EEG-Based Emotion Recognition

Electroencephalography (EEG) based emotion recognition enables machines to perceive users' affective states, which has attracted increasing attention. However, most of the current emotion recognition methods neglect the structural information among different brain regions, which can lead to the incorrect learning of high-level EEG feature representation. To mitigate possible performance degradation, we propose a novel nuclear norm regularized deep neural network framework (NRDNN) that can capture the structural information among different brain regions in EEG decoding. The proposed NRDNN first utilizes deep neural networks to learn high-level feature representations of multiple brain regions, respectively. Then, a set of weights indicating the contributions of each brain region can be automatically learned using a region-attention layer. Subsequently, the weighted feature representations of multiple brain regions are stacked into a feature matrix, and the nuclear norm regularization is adopted to learn the structural information within the feature matrix. The proposed NRDNN method can learn the high-level representations of EEG signals within multiple brain regions, and the contributions of them can be automatically adjusted by assigning a set of weights. Besides, the structural information among multiple brain regions can be captured in the learning procedure. Finally, the proposed NRDNN can perform in an efficient end-to-end manner. We conducted extensive experiments on publicly available emotion EEG dataset to evaluate the effectiveness of the proposed NRDNN. Experimental results demonstrated that the proposed NRDNN can achieve state-of-the-art performance by leveraging the structural information.

Transfer learning for motor imagery based brain-computer interfaces: A tutorial

A brain-computer interface (BCI) enables a user to communicate directly with an external device, e.g., a computer, using brain signals. It can be used to research, map, assist, augment, or repair human cognitive or sensory-motor functions. A closed-loop BCI system performs signal acquisition, temporal filtering, spatial filtering, feature engineering and classification, before sending out the control signal to an external device. Transfer learning (TL) has been widely used in motor imagery (MI) based BCIs to reduce the calibration effort for a new subject, greatly increasing their utility. This tutorial describes how TL can be considered in as many components of a BCI system as possible, and introduces a complete TL pipeline for MI-based BCIs. Examples on two MI datasets demonstrated the advantages of considering TL in multiple components of MI-based BCIs. Especially, integrating data alignment and sophisticated TL approaches can significantly improve the classification performance, and hence greatly reduces the calibration effort.

A Domain Adaptation-Based Method for Classification of Motor Imagery EEG

Non-stationarity of EEG signals lead to high variability across sessions, which results in low classification accuracy. To reduce the inter-session variability, an unsupervised domain adaptation method is proposed. Arithmetic mean and covariance are exploited to represent the data distribution. First, overall mean alignment is conducted between the source and target data. Then, the data in the target domain is labeled by a classifier trained with the source data. The per-class mean and covariance of the target data are estimated based on the predicted labels. Next, an alignment from the source domain to the target domain is performed according to the covariance of each class in the target domain. Finally, per-class mean adaptation is required after covariance alignment to remove the shift of data distribution caused by covariance alignment. Two public BCI competition datasets, namely the BCI competition III dataset IVa and the BCI competition IV dataset IIa were used to evaluate the proposed method. On both datasets, the proposed method effectively improved classification accuracy.

Attention-Based DSC-ConvLSTM for Multiclass Motor Imagery Classification

With the rapid development of deep learning, researchers have gradually applied it to motor imagery brain computer interface (MI-BCI) and initially demonstrated its advantages over traditional machine learning. However, its application still faces many challenges, and the recognition rate of electroencephalogram (EEG) is still the bottleneck restricting the development of MI-BCI. In order to improve the accuracy of EEG classification, a DSC-ConvLSTM model based on the attention mechanism is proposed for the multi-classification of motor imagery EEG signals. To address the problem of the small sample size of well-labeled and accurate EEG data, the preprocessing uses sliding windows for data augmentation, and the average prediction loss of each sliding window is used as the final prediction loss for that trial. This not only increases the training sample size and is beneficial to train complex neural network models, but also the network no longer extracts the global features of the whole trial so as to avoid learning the difference features among trials, which can effectively eliminate the influence of individual specificity. In the aspect of feature extraction and classification, the overall network structure is designed according to the characteristics of the EEG signals in this paper. Firstly, depth separable convolution (DSC) is used to extract spatial features of EEG signals. On the one hand, this reduces the number of parameters and improves the response speed of the system. On the other hand, the network structure we designed is more conducive to extract directly the direct extraction of spatial features of EEG signals. Secondly, the internal structure of the Long Short-Term Memory (LSTM) unit is improved by using convolution and attention mechanism, and a novel bidirectional convolution LSTM (ConvLSTM) structure is proposed by comparing the effects of embedding convolution and attention mechanism in the input and different gates, respectively. In the ConvLSTM module, the convolutional structure is only introduced into the input-to-state transition, while the gates still remain the original fully connected mechanism, and the attention mechanism is introduced into the input to further improve the overall decoding performance of the model. This bidirectional ConvLSTM extracts the time-domain features of EEG signals and integrates the feature extraction capability of the CNN and the sequence processing capability of LSTM. The experimental results show that the average classification accuracy of the model reaches 73.7% and 92.6% on two datasets, BCI Competition IV Dataset 2a and High Gamma Dataset, respectively, which proves the robustness and effectiveness of the model we proposed. It can be seen that the model in this paper can deeply excavate significant EEG features from the original EEG signals, show good performance in different subjects and different datasets, and improve the influence of individual variability on the classification performance, which is of practical significance for promoting the development of brain-computer interface technology towards a practical and marketable direction.

Low-quality Training Data Detection Method of EEG Signals for Motor Imagery BCI System

BACKGROUND

The design and implementation of high-performance motor imagery-based brain computer interface (MI-BCI) requires high-quality training samples. However, fluctuation in subjects' physiological and mental states as well as artifacts can produce the low-quality motor imagery electroencephalogram (EEG) signal, which will damage the performance of MI-BCI system.

NEW METHOD

In order to select high-quality MI-EEG training data, this paper proposes a low-quality training data detection method combining independent component analysis (ICA) and weak classifier cluster. we also design and implement a new online BCI system based on motor imagery to verify the online processing performance of the proposed method.

RESULT

In order to verify the effectiveness of the proposed method, we conducted offline experiments on the public dataset called BCI Competition IV Data Set 2b. Furthermore, in order to verify the processing performance of the online system, we designed 60 groups of online experiments on 12 subjects. The online experimental results show that the twelve subjects can complete the system task efficiently (the best experiment is 135.6 seconds with 9 trials of subject S1).

CONCLUSION

This paper demonstrated that the proposed low-quality training data detection method can effectively screen out low-quality training samples, so as to improve the performance of the MI-BCI system.

Robustly Effective Approaches on Motor Imagery-Based Brain Computer Interfaces

Motor Imagery Brain Computer Interfaces (MI-BCIs) are systems that receive the users’ brain activity as an input signal in order to communicate between the brain and the interface or an action to be performed through the detection of the imagination of a movement. Brainwaves’ features are crucial for the performance of the interface to be increased. The robustness of these features must be ensured in order for the effectiveness to remain high in various subjects. The present work consists of a review, which includes scientific publications related to the use of robust feature extraction methods in Motor Imagery from 2017 until today. The research showed that the majority of the works focus on spatial features through Common Spatial Patterns (CSP) methods (44.26%). Based on the combination of accuracy percentages and K-values, which show the effectiveness of each approach, Wavelet Transform (WT) has shown higher robustness than CSP and PSD methods in the majority of the datasets used for comparison and also in the majority of the works included in the present review, although they had a lower usage percentage in the literature (16.65%). The research showed that there was an increase in 2019 of the detection of spatial features to increase the robustness of an approach, but the time-frequency features, or a combination of those, achieve better results with their increase starting from 2019 onwards. Additionally, Wavelet Transforms and their variants, in combination with deep learning, manage to achieve high percentages thus making a method robustly accurate.

Analytic beamformer transformation for transfer learning in motion-onset visual evoked potential decoding

OBJECTIVE

While decoders of EEG-based event-related potentials (ERPs) are routinely tailored to the individual user to maximize performance, developing them on populations for individual usage has proven much more challenging. We propose the analytic beamformer transformation (ABT) to extract phase and/or magnitude information from spatiotemporal ERPs in response to motion-onset stimulation.

APPROACH

We have tested ABT on 52 motion-onset visual evoked potential (mVEP) datasets from 26 healthy subjects and compared the classification accuracy of support vector machine (SVM), spatiotemporal beamformer (stBF) and stepwise linear discriminant analysis (SWLDA) when trained on individual subjects and on a population thereof.

MAIN RESULTS

When using phase- and combined phase/magnitude information extracted by ABT, we show significant improvements in accuracy of population-trained classifiers applied to individual users (p<0.001). We also show that 450 epochs are needed for a correct functioning of ABT, which corresponds to 2 minutes of paradigm stimulation.

SIGNIFICANCE

We have shown that ABT can be used to create population-trained mVEP classifiers using a limited number of epochs. We expect this to pertain to other ERPs or synchronous stimulation paradigms, allowing for a more effective, population-based training of visual BCIs. Finally, as ABT renders recordings across subjects more structurally invariant, it could be used for transfer learning purposes in view of plug-and-play BCI applications.