Optimizing spatial filters by minimizing within-class dissimilarities in electroencephalogram-based brain-computer interface

Exploring Feature Selection and Classification Techniques to Improve the Performance of an Electroencephalography-Based Motor Imagery Brain-Computer Interface System

The accuracy of classifying motor imagery (MI) activities is a significant challenge when using brain-computer interfaces (BCIs). BCIs allow people with motor impairments to control external devices directly with their brains using electroencephalogram (EEG) patterns that translate brain activity into control signals. Many researchers have been working to develop MI-based BCI recognition systems using various time-frequency feature extraction and classification approaches. However, the existing systems still face challenges in achieving satisfactory performance due to large amount of non-discriminative and ineffective features. To get around these problems, we suggested a multiband decomposition-based feature extraction and classification method that works well, along with a strong feature selection method for MI tasks. Our method starts by splitting the preprocessed EEG signal into four sub-bands. In each sub-band, we then used a common spatial pattern (CSP) technique to pull out narrowband-oriented useful features, which gives us a high-dimensional feature vector. Subsequently, we utilized an effective feature selection method, Relief-F, which reduces the dimensionality of the final features. Finally, incorporating advanced classification techniques, we classified the final reduced feature vector. To evaluate the proposed model, we used the three different EEG-based MI benchmark datasets, and our proposed model achieved better performance accuracy than existing systems. Our model's strong points include its ability to effectively reduce feature dimensionality and improve classification accuracy through advanced feature extraction and selection methods.

Adaptive Multimodel Knowledge Transfer Matrix Machine for EEG Classification

The emerging matrix learning methods have achieved promising performances in electroencephalogram (EEG) classification by exploiting the structural information between the columns or rows of feature matrices. Due to the intersubject variability of EEG data, these methods generally need to collect a large amount of labeled individual EEG data, which would cause fatigue and inconvenience to the subjects. Insufficient subject-specific EEG data will weaken the generalization capability of the matrix learning methods in neural pattern decoding. To overcome this dilemma, we propose an adaptive multimodel knowledge transfer matrix machine (AMK-TMM), which can selectively leverage model knowledge from multiple source subjects and capture the structural information of the corresponding EEG feature matrices. Specifically, by incorporating least-squares (LS) loss with spectral elastic net regularization, we first present an LS support matrix machine (LS-SMM) to model the EEG feature matrices. To boost the generalization capability of LS-SMM in scenarios with limited EEG data, we then propose a multimodel adaption method, which can adaptively choose multiple correlated source model knowledge with a leave-one-out cross-validation strategy on the available target training data. We extensively evaluate our method on three independent EEG datasets. Experimental results demonstrate that our method achieves promising performances on EEG classification.

EEG Channel Selection Techniques in Motor Imagery Applications: A Review and New Perspectives

Communication, neuro-prosthetics, and environmental control are just a few applications for disabled persons who use robots and manipulators that use brain-computer interface (BCI) systems. The brain's motor imagery (MI) signal is an essential input for a brain-related task in BCI applications. Due to their noninvasive, portability, and cost-effectiveness, electroencephalography (EEG) signals are the most widely used input in BCI systems. The EEG data are often collected from more than 100 different locations in the brain; channel selection techniques are critical for selecting the optimum channels for a given application. However, when analyzing EEG data, the principal purpose of channel selection is to reduce computational complexity, improve classification accuracy by avoiding overfitting, and reduce setup time. Several channel selection assessment algorithms, both with and without classification-based methods, extracted appropriate channel subsets using defined criteria. Therefore, based on the exhaustive analysis of the EEG channel selection, this manuscript analyses several existing studies to reduce the number of noisy channels and improve system performance. We review several existing works to find the most promising MI-based EEG channel selection algorithms and associated classification methodologies on various datasets. Moreover, we focus on channel selection methods that choose fewer channels with great precision. Finally, our main finding is that a smaller channel set, typically 10-30% of total channels, provided excellent performance compared to other existing studies.

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.

A Transfer Learning Algorithm to Reduce Brain-Computer Interface Calibration Time for Long-Term Users

Current motor imagery-based brain-computer interface (BCI) systems require a long calibration time at the beginning of each session before they can be used with adequate levels of classification accuracy. In particular, this issue can be a significant burden for long term BCI users. This article proposes a novel transfer learning algorithm, called r-KLwDSA, to reduce the BCI calibration time for long-term users. The proposed r-KLwDSA algorithm aligns the user's EEG data collected in previous sessions to the few EEG trials collected in the current session, using a novel linear alignment method. Thereafter, the aligned EEG trials from the previous sessions and the few EEG trials from the current sessions are fused through a weighting mechanism before they are used for calibrating the BCI model. To validate the proposed algorithm, a large dataset containing the EEG data from 11 stroke patients, each performing 18 BCI sessions, was used. The proposed framework demonstrated a significant improvement in the classification accuracy, of over 4% compared to the session-specific algorithm, when there were as few as two trials per class available from the current session. The proposed algorithm was particularly successful in improving the BCI accuracy of the sessions that had initial session-specific accuracy below 60%, with an average improvement of around 10% in the accuracy, leading to more stroke patients having meaningful BCI rehabilitation.

Multiscale time-frequency method for multiclass Motor Imagery Brain Computer Interface

Motor Imagery Brain Computer Interface (MI-BCI) has become a promising technology in the field of neurorehabilitation. However, the performance and computational complexity of the current multiclass MI-BCI have not been fully optimized, and the intuitive interpretation of individual differences on motor imagery tasks is seldom investigated. In this paper, a well-designed multiscale time-frequency segmentation scheme is first applied to multichannel EEG recordings to obtain Time-Frequency Segments (TFSs). Then, the TFS selection based on a specific wrapper feature selection rule is utilized to determine optimum TFSs. Next, One-Versus-One (OvO)-divCSP implemented in divergence framework is used to extract discriminative features. Finally, One-Versus-Rest (OvR)-SVM is utilized to predict the class label based on selected multiclass MI features. Experimental results indicate our method yields a superior performance on two publicly available multiclass MI datasets with a mean accuracy of 80.00% and a mean kappa of 0.73. Meanwhile, the proposed TFS selection method can significantly alleviate the computational burden with little accuracy reduction, demonstrating the feasibility of real-time multiclass MI-BCI. Furthermore, the Motor Imagery Time-Frequency Reaction Map (MI-TFRM) is visualized, contributing to analyzing and interpreting the performance differences between different subjects.

Internal Feature Selection Method of CSP Based on L1-Norm and Dempster-Shafer Theory

The common spatial pattern (CSP) algorithm is a well-recognized spatial filtering method for feature extraction in motor imagery (MI)-based brain-computer interfaces (BCIs). However, due to the influence of nonstationary in electroencephalography (EEG) and inherent defects of the CSP objective function, the spatial filters, and their corresponding features are not necessarily optimal in the feature space used within CSP. In this work, we design a new feature selection method to address this issue by selecting features based on an improved objective function. Especially, improvements are made in suppressing outliers and discovering features with larger interclass distances. Moreover, a fusion algorithm based on the Dempster-Shafer theory is proposed, which takes into consideration the distribution of features. With two competition data sets, we first evaluate the performance of the improved objective functions in terms of classification accuracy, feature distribution, and embeddability. Then, a comparison with other feature selection methods is carried out in both accuracy and computational time. Experimental results show that the proposed methods consume less additional computational cost and result in a significant increase in the performance of MI-based BCI systems.

An Inter- and Intra-Subject Transfer Calibration Scheme for Improving Feedback Performance of Sensorimotor Rhythm-Based BCI Rehabilitation

The Brain Computer Interface (BCI) system is a typical neurophysiological application which helps paralyzed patients with human-machine communication. Stroke patients with motor disabilities are able to perform BCI tasks for clinical rehabilitation. This paper proposes an effective scheme of transfer calibration for BCI rehabilitation. The inter- and intra-subject transfer learning approaches can improve the low-precision classification performance for experimental feedback. The results imply that the systematical scheme is positive in increasing the confidence of voluntary training for stroke patients. In addition, it also reduces the time consumption of classifier calibration.

Mitigation of ocular artifacts for EEG signal using improved earth worm optimization-based neural network and lifting wavelet transform

An Electroencephalogram (EEG) is often tarnished by various categories of artifacts. Numerous efforts have been taken to improve its quality by eliminating the artifacts. The EEG involves the biological artifacts (ocular artifacts, ECG and EMG artifacts), and technical artifacts (noise from the electric power source, amplitude artifacts, etc.). From these physiological artifacts, ocular activities are one of the most well-known over other noise sources. Reducing the risks of this event and avoid it is practically very difficult, even impossible, as the ocular activities are involuntary tasks. To trim down the effect of ocular artifacts overlapping with EEG signal and overwhelm the subjected flaws, few intelligent approaches have to be developed. This proposal tempts to implement a novel method for detecting and preventing ocular artifacts from the EEG signal. The developed model involves two main phases: (a) Detection of Ocular artifacts and (b) Removal of ocular artifacts. For detecting the ocular artifacts, initially, the EEG is subjected to decomposition process using 5-level Discrete Wavelet Transform (DWT), and Empirical Mean Curve Decomposition (EMCD). Next to the decomposition process, the features like kurtosis, variance, Shannon's entropy, and few first-order statistical features are extracted. These features will be helpful for the detection process in the classification side. For detecting the ocular artifacts from the decomposed signal, the extracted features are subjected to a machine learning algorithm called Neural Network (NN). As an improvement to the conventional NN, the training algorithm of ANN is improved by the improved Earth Worm optimization Algorithm (EWA) termed as Dual Positioned Elitism-based EWA (DPE-EWA), which updates the weight of NN to improve the performance. In the Removal phase, the optimized Lifting Wavelet Transform (LWT) is deployed, in which the improvement is made on optimizing the filter coefficients using the proposed DPE-EWA. Thus, the integration of optimized NN and optimized LWT suggests a potential possibility to accommodate the detection and removal of ocular artifacts that exist in the EEG signals.

Neural component analysis: A spatial filter for electroencephalogram analysis

BACKGROUND

Spatial filtering and source separation are valuable tools in the analysis of EEG data. However, despite the well-known spatial localisation of individual cognitive processes within the brain, the available methods for source separation, such as the widely used blind source separation technique, do not take into account the spatial distributions and locations of sources. This can result in sub-optimal source identification.

NEW METHOD

We present a new method for deriving a spatial filter for EEG data that attempts to identify sources that are maximally spatially distinct from one another in terms of the spatial distributions of their projections.

RESULTS

We first evaluate our method with simulated EEG and show that it is able to separate EEG signals into components with distinct spatial distributions that closely resemble the original simulated sources. We also evaluate our method with real EEG and show it is able to identify a spatial filter that can be used to significantly improve classification accuracy of the P300 event-related potential (ERP).

COMPARISON WITH EXISTING METHODS

We compare our method to a state of the art blind source separation methods, fast independent component analysis (ICA) and common spatial patterns (CSP). We evaluate the methods suitability for a common source separation application, analysis of ERPs.

CONCLUSIONS

Our results show that our method is well suited to identifying spatial filters for EEG analysis. This has potential applications in a wide range of EEG signal processing applications.

Assessing impact of channel selection on decoding of motor and cognitive imagery from MEG data

OBJECTIVE

Magnetoencephalography (MEG) based brain-computer interface (BCI) involves a large number of sensors allowing better spatiotemporal resolution for assessing brain activity patterns. There have been many efforts to develop BCI using MEG with high accuracy, though an increase in the number of channels (NoC) means an increase in computational complexity. However, not all sensors necessarily contribute significantly to an increase in classification accuracy (CA) and specifically in the case of MEG-based BCI no channel selection methodology has been performed. Therefore, this study investigates the effect of channel selection on the performance of MEG-based BCI.

APPROACH

MEG data were recorded for two sessions from 15 healthy participants performing motor imagery, cognitive imagery and a mixed imagery task pair using a unique paradigm. Performance of four state-of-the-art channel selection methods (i.e. Class-Correlation, ReliefF, Random Forest, and Infinite Latent Feature Selection were applied across six binary tasks in three different frequency bands) were evaluated in this study on two state-of-the-art features, i.e. bandpower and common spatial pattern (CSP).

MAIN RESULTS

All four methods provided a statistically significant increase in CA compared to a baseline method using all gradiometer sensors, i.e. 204 channels with band-power features from alpha (8-12 Hz), beta (13-30 Hz), or broadband (α + β) (8-30 Hz). It is also observed that the alpha frequency band performed better than the beta and broadband frequency bands. The performance of the beta band gave the lowest CA compared with the other two bands. Channel selection improved accuracy irrespective of feature types. Moreover, all the methods reduced the NoC significantly, from 204 to a range of 1-25, using bandpower as a feature and from 15 to 105 for CSP. The optimal channel number also varied not only in each session but also for each participant. Reducing the NoC will help to decrease the computational cost and maintain numerical stability in cases of low trial numbers.

SIGNIFICANCE

The study showed significant improvement in performance of MEG-BCI with channel selection irrespective of feature type and hence can be successfully applied for BCI applications.

Adaptive feature extraction in EEG-based motor imagery BCI: tracking mental fatigue

OBJECTIVE

Electroencephalogram (EEG) signals are non-stationary. This could be due to internal fluctuation of brain states such as fatigue, frustration, etc. This necessitates the development of adaptive brain-computer interfaces (BCI) whose performance does not deteriorate significantly with the adversary change in the cognitive state. In this paper, we put forward an unsupervised adaptive scheme to adapt the feature extractor of motor imagery (MI) BCIs by tracking the fatigue level of the user.

APPROACH

Eleven subjects participated in the study during which they accomplished MI tasks while self-reporting their perceived levels of mental fatigue. Out of the 11 subjects, only six completed the whole experiment, while the others quit in the middle because of experiencing high fatigue. The adaptive feature extractor is attained through the adaptation of the common spatial patterns (CSP), one of the most popular feature extraction algorithms in EEG-based BCIs. The proposed method was analyzed in two ways: offline and in near real-time. The separability of the MI EEG features extracted by the proposed adaptive CSP (ADCSP) has been compared with that by the conventional CSP (C-CSP) and another CSP based adaptive method (ACSP) in terms of: Davies Bouldin index (DBI), Fisher score (FS) and Dunn's index (DI).

MAIN RESULTS

Experimental results show significant improvement in the separability of MI EEG features extracted by ADCSP as compared to that by C-CSP and ACSP.

SIGNIFICANCE

Collectively, the results of the experiments in this study suggest that adapting CSP based on mental fatigue can improve the class separability of MI EEG features.

Characterizing Regularization Techniques for Spatial Filter Optimization in Oscillatory EEG Regression Problems : Guidelines Derived from Simulation and Real-World Data

We report on novel supervised algorithms for single-trial brain state decoding. Their reliability and robustness are essential to efficiently perform neurotechnological applications in closed-loop. When brain activity is assessed by multichannel recordings, spatial filters computed by the source power comodulation (SPoC) algorithm allow identifying oscillatory subspaces. They regress to a known continuous trial-wise variable reflecting, e.g. stimulus characteristics, cognitive processing or behavior. In small dataset scenarios, this supervised method tends to overfit to its training data as the involved recordings via electroencephalogram (EEG), magnetoencephalogram or local field potentials generally provide a low signal-to-noise ratio. To improve upon this, we propose and characterize three types of regularization techniques for SPoC: approaches using Tikhonov regularization (which requires model selection via cross-validation), combinations of Tikhonov regularization and covariance matrix normalization as well as strategies exploiting analytical covariance matrix shrinkage. All proposed techniques were evaluated both in a novel simulation framework and on real-world data. Based on the simulation findings, we saw our expectations fulfilled, that SPoC regularization generally reveals the largest benefit for small training sets and under severe label noise conditions. Relevant for practitioners, we derived operating ranges of regularization hyperparameters for cross-validation based approaches and offer open source code. Evaluating all methods additionally on real-world data, we observed an improved regression performance mainly for datasets from subjects with initially poor performance. With this proof-of-concept paper, we provided a generalizable regularization framework for SPoC which may serve as a starting point for implementing advanced techniques in the future.

Covariate shift estimation based adaptive ensemble learning for handling non-stationarity in motor imagery related EEG-based brain-computer interface

The non-stationary nature of electroencephalography (EEG) signals makes an EEG-based brain-computer interface (BCI) a dynamic system, thus improving its performance is a challenging task. In addition, it is well-known that due to non-stationarity based covariate shifts, the input data distributions of EEG-based BCI systems change during inter- and intra-session transitions, which poses great difficulty for developments of online adaptive data-driven systems. Ensemble learning approaches have been used previously to tackle this challenge. However, passive scheme based implementation leads to poor efficiency while increasing high computational cost. This paper presents a novel integration of covariate shift estimation and unsupervised adaptive ensemble learning (CSE-UAEL) to tackle non-stationarity in motor-imagery (MI) related EEG classification. The proposed method first employs an exponentially weighted moving average model to detect the covariate shifts in the common spatial pattern features extracted from MI related brain responses. Then, a classifier ensemble was created and updated over time to account for changes in streaming input data distribution wherein new classifiers are added to the ensemble in accordance with estimated shifts. Furthermore, using two publicly available BCI-related EEG datasets, the proposed method was extensively compared with the state-of-the-art single-classifier based passive scheme, single-classifier based active scheme and ensemble based passive schemes. The experimental results show that the proposed active scheme based ensemble learning algorithm significantly enhances the BCI performance in MI classifications.

Classification of electroencephalogram signals using wavelet-CSP and projection extreme learning machine

Brain-computer interface (BCI) systems establish a direct communication channel from the brain to an output device. As the basis of BCIs, recognizing motor imagery activities poses a considerable challenge to signal processing due to the complex and non-stationary characteristics. This paper introduces an optimal and intelligent method for motor imagery BCIs. Because of the robustness to noise, wavelet packet decomposition and common spatial pattern (CSP) methods were implemented to reduce the dimensions of preprocessed signals. And a novel and efficient classifier projection extreme learning machine (PELM) was employed to recognize the labels of electroencephalogram signals. Experiments have been performed on the BCI Competition Dataset to demonstrate the superiority of wavelet-CSP in BCI and the outperformance of the PELM-based method. Results show that the average recognition rate of PELM approaches approximately 70%, while the optimal rate of other methods is 72%, whose training time and classification time are relatively longer as 11.00 ms and 11.66 ms, respectively, compared with 4.75 ms and 4.87 ms obtained by using the proposed BCI system.

Information Theoretic Approaches for Motor-Imagery BCI Systems: Review and Experimental Comparison

Brain computer interfaces (BCIs) have been attracting a great interest in recent years. The common spatial patterns (CSP) technique is a well-established approach to the spatial filtering of the electroencephalogram (EEG) data in BCI applications. Even though CSP was originally proposed from a heuristic viewpoint, it can be also built on very strong foundations using information theory. This paper reviews the relationship between CSP and several information-theoretic approaches, including the Kullback-Leibler divergence, the Beta divergence and the Alpha-Beta log-det (AB-LD)divergence. We also revise other approaches based on the idea of selecting those features that are maximally informative about the class labels. The performance of all the methods will be also compared via experiments.

A Unified Fisher's Ratio Learning Method for Spatial Filter Optimization

To detect the mental task of interest, spatial filtering has been widely used to enhance the spatial resolution of electroencephalography (EEG). However, the effectiveness of spatial filtering is undermined due to the significant nonstationarity of EEG. Based on regularization, most of the conventional stationary spatial filter design methods address the nonstationarity at the cost of the interclass discrimination. Moreover, spatial filter optimization is inconsistent with feature extraction when EEG covariance matrices could not be jointly diagonalized due to the regularization. In this paper, we propose a novel framework for a spatial filter design. With Fisher's ratio in feature space directly used as the objective function, the spatial filter optimization is unified with feature extraction. Given its ratio form, the selection of the regularization parameter could be avoided. We evaluate the proposed method on a binary motor imagery data set of 16 subjects, who performed the calibration and test sessions on different days. The experimental results show that the proposed method yields improvement in classification performance for both single broadband and filter bank settings compared with conventional nonunified methods. We also provide a systematic attempt to compare different objective functions in modeling data nonstationarity with simulation studies.To detect the mental task of interest, spatial filtering has been widely used to enhance the spatial resolution of electroencephalography (EEG). However, the effectiveness of spatial filtering is undermined due to the significant nonstationarity of EEG. Based on regularization, most of the conventional stationary spatial filter design methods address the nonstationarity at the cost of the interclass discrimination. Moreover, spatial filter optimization is inconsistent with feature extraction when EEG covariance matrices could not be jointly diagonalized due to the regularization. In this paper, we propose a novel framework for a spatial filter design. With Fisher's ratio in feature space directly used as the objective function, the spatial filter optimization is unified with feature extraction. Given its ratio form, the selection of the regularization parameter could be avoided. We evaluate the proposed method on a binary motor imagery data set of 16 subjects, who performed the calibration and test sessions on different days. The experimental results show that the proposed method yields improvement in classification performance for both single broadband and filter bank settings compared with conventional nonunified methods. We also provide a systematic attempt to compare different objective functions in modeling data nonstationarity with simulation studies.

Use of Electroencephalography Brain-Computer Interface Systems as a Rehabilitative Approach for Upper Limb Function After a Stroke: A Systematic Review

BACKGROUND

Brain-computer interface (BCI) systems have been suggested as a promising tool for neurorehabilitation. However, to date, there is a lack of homogeneous findings. Furthermore, no systematic reviews have analyzed the degree of validation of these interventions for upper limb (UL) motor rehabilitation poststroke.

OBJECTIVES

The study aims were to compile all available studies that assess an UL intervention based on an electroencephalography (EEG) BCI system in stroke; to analyze the methodological quality of the studies retrieved; and to determine the effects of these interventions on the improvement of motor abilities. TYPE: This was a systematic review.

LITERATURE SURVEY

Searches were conducted in PubMed, PEDro, Embase, Cumulative Index to Nursing and Allied Health, Web of Science, and Cochrane Central Register of Controlled Trial from inception to September 30, 2015.

METHODOLOGY

This systematic review compiles all available studies that assess UL intervention based on an EEG-BCI system in patients with stroke, analyzing their methodological quality using the Critical Review Form for Quantitative Studies, and determining the grade of recommendation of these interventions for improving motor abilities as established by the Oxford Centre for Evidence-based Medicine. The articles were selected according to the following criteria: studies evaluating an EEG-based BCI intervention; studies including patients with a stroke and hemiplegia, regardless of lesion origin or temporal evolution; interventions using an EEG-based BCI to restore functional abilities of the affected UL, regardless of the interface used or its combination with other therapies; and studies using validated tools to evaluate motor function.

SYNTHESIS

After the literature search, 13 articles were included in this review: 4 studies were randomized controlled trials; 1 study was a controlled study; 4 studies were case series studies; and 4 studies were case reports. The methodological quality of the included papers ranged from 6 to 15, and the level of evidence varied from 1b to 5. The articles included in this review involved a total of 141 stroke patients.

CONCLUSIONS

This systematic review suggests that BCI interventions may be a promising rehabilitation approach in subjects with stroke.

LEVEL OF EVIDENCE

II.

A spatial-frequency-temporal optimized feature sparse representation-based classification method for motor imagery EEG pattern recognition

Effective feature extraction and classification methods are of great importance for motor imagery (MI)-based brain-computer interface (BCI) systems. The common spatial pattern (CSP) algorithm is a widely used feature extraction method for MI-based BCIs. In this work, we propose a novel spatial-frequency-temporal optimized feature sparse representation-based classification method. Optimal channels are selected based on relative entropy criteria. Significant CSP features on frequency-temporal domains are selected automatically to generate a column vector for sparse representation-based classification (SRC). We analyzed the performance of the new method on two public EEG datasets, namely BCI competition III dataset IVa which has five subjects and BCI competition IV dataset IIb which has nine subjects. Compared to the performance offered by the existing SRC method, the proposed method achieves average classification accuracy improvements of 21.568 and 14.38% for BCI competition III dataset IVa and BCI competition IV dataset IIb, respectively. Furthermore, our approach also shows better classification performance when compared to other competing methods for both datasets.

Sparse Bayesian Classification of EEG for Brain-Computer Interface

Regularization has been one of the most popular approaches to prevent overfitting in electroencephalogram (EEG) classification of brain-computer interfaces (BCIs). The effectiveness of regularization is often highly dependent on the selection of regularization parameters that are typically determined by cross-validation (CV). However, the CV imposes two main limitations on BCIs: 1) a large amount of training data is required from the user and 2) it takes a relatively long time to calibrate the classifier. These limitations substantially deteriorate the system's practicability and may cause a user to be reluctant to use BCIs. In this paper, we introduce a sparse Bayesian method by exploiting Laplace priors, namely, SBLaplace, for EEG classification. A sparse discriminant vector is learned with a Laplace prior in a hierarchical fashion under a Bayesian evidence framework. All required model parameters are automatically estimated from training data without the need of CV. Extensive comparisons are carried out between the SBLaplace algorithm and several other competing methods based on two EEG data sets. The experimental results demonstrate that the SBLaplace algorithm achieves better overall performance than the competing algorithms for EEG classification.

Optimizing Single-Trial EEG Classification by Stationary Matrix Logistic Regression in Brain-Computer Interface

In addition to the noisy and limited spatial resolution characteristics of the electroencephalography (EEG) signal, the intrinsic nonstationarity in the EEG data makes the single-trial EEG classification an even more challenging problem in brain-computer interface (BCI). Variations of the signal properties within a session often result in deteriorated classification performance. This is mainly attributed to the reason that the routine feature extraction or classification method does not take the changes in the signal into account. Although several extensions to the standard feature extraction method have been proposed to reduce the sensitivity to nonstationarity in data, they optimize different objective functions from that of the subsequent classification model, and thereby, the extracted features may not be optimized for the classification. In this paper, we propose an approach that directly optimizes the classifier's discriminativity and robustness against the within-session nonstationarity of the EEG data through a single optimization paradigm, and show that it can greatly improve the performance, in particular for the subjects who have difficulty in controlling a BCI. Moreover, the experimental results on two benchmark data sets demonstrate that our approach significantly outperforms the compared approaches in reducing classification error rates.

Improving the discrimination of hand motor imagery via virtual reality based visual guidance

While research on the brain-computer interface (BCI) has been active in recent years, how to get high-quality electrical brain signals to accurately recognize human intentions for reliable communication and interaction is still a challenging task. The evidence has shown that visually guided motor imagery (MI) can modulate sensorimotor electroencephalographic (EEG) rhythms in humans, but how to design and implement efficient visual guidance during MI in order to produce better event-related desynchronization (ERD) patterns is still unclear. The aim of this paper is to investigate the effect of using object-oriented movements in a virtual environment as visual guidance on the modulation of sensorimotor EEG rhythms generated by hand MI. To improve the classification accuracy on MI, we further propose an algorithm to automatically extract subject-specific optimal frequency and time bands for the discrimination of ERD patterns produced by left and right hand MI. The experimental results show that the average classification accuracy of object-directed scenarios is much better than that of non-object-directed scenarios (76.87% vs. 69.66%). The result of the t-test measuring the difference between them is statistically significant (p = 0.0207). When compared to algorithms based on fixed frequency and time bands, contralateral dominant ERD patterns can be enhanced by using the subject-specific optimal frequency and the time bands obtained by our proposed algorithm. These findings have the potential to improve the efficacy and robustness of MI-based BCI applications.

Comparison Analysis: Granger Causality and New Causality and Their Applications to Motor Imagery

In this paper we first point out a fatal drawback that the widely used Granger causality (GC) needs to estimate the autoregressive model, which is equivalent to taking a series of backward recursive operations which are infeasible in many irreversible chemical reaction models. Thus, new causality (NC) proposed by Hu et al. (2011) is theoretically shown to be more sensitive to reveal true causality than GC. We then apply GC and NC to motor imagery (MI) which is an important mental process in cognitive neuroscience and psychology and has received growing attention for a long time. We study causality flow during MI using scalp electroencephalograms from nine subjects in Brain-computer interface competition IV held in 2008. We are interested in three regions: Cz (central area of the cerebral cortex), C3 (left area of the cerebral cortex), and C4 (right area of the cerebral cortex) which are considered to be optimal locations for recognizing MI states in the literature. Our results show that: 1) there is strong directional connectivity from Cz to C3/C4 during left- and right-hand MIs based on GC and NC; 2) during left-hand MI, there is directional connectivity from C4 to C3 based on GC and NC; 3) during right-hand MI, there is strong directional connectivity from C3 to C4 which is much clearly revealed by NC than by GC, i.e., NC largely improves the classification rate; and 4) NC is demonstrated to be much more sensitive to reveal causal influence between different brain regions than GC.

Subject-based feature extraction by using fisher WPD-CSP in brain-computer interfaces

BACKGROUND AND OBJECTIVE

Feature extraction of electroencephalogram (EEG) plays a vital role in brain-computer interfaces (BCIs). In recent years, common spatial pattern (CSP) has been proven to be an effective feature extraction method. However, the traditional CSP has disadvantages of requiring a lot of input channels and the lack of frequency information. In order to remedy the defects of CSP, wavelet packet decomposition (WPD) and CSP are combined to extract effective features. But WPD-CSP method considers less about extracting specific features that are fitted for the specific subject. So a subject-based feature extraction method using fisher WPD-CSP is proposed in this paper.

METHODS

The idea of proposed method is to adapt fisher WPD-CSP to each subject separately. It mainly includes the following six steps: (1) original EEG signals from all channels are decomposed into a series of sub-bands using WPD; (2) average power values of obtained sub-bands are computed; (3) the specified sub-bands with larger values of fisher distance according to average power are selected for that particular subject; (4) each selected sub-band is reconstructed to be regarded as a new EEG channel; (5) all new EEG channels are used as input of the CSP and a six-dimensional feature vector is obtained by the CSP. The subject-based feature extraction model is so formed; (6) the probabilistic neural network (PNN) is used as the classifier and the classification accuracy is obtained.

RESULTS

Data from six subjects are processed by the subject-based fisher WPD-CSP, the non-subject-based fisher WPD-CSP and WPD-CSP, respectively. Compared with non-subject-based fisher WPD-CSP and WPD-CSP, the results show that the proposed method yields better performance (sensitivity: 88.7±0.9%, and specificity: 91±1%) and the classification accuracy from subject-based fisher WPD-CSP is increased by 6-12% and 14%, respectively.

CONCLUSIONS

The proposed subject-based fisher WPD-CSP method can not only remedy disadvantages of CSP by WPD but also discriminate helpless sub-bands for each subject and make remaining fewer sub-bands keep better separability by fisher distance, which leads to a higher classification accuracy than WPD-CSP method.

Optimizing spatial patterns with sparse filter bands for motor-imagery based brain-computer interface

BACKGROUND

Common spatial pattern (CSP) has been most popularly applied to motor-imagery (MI) feature extraction for classification in brain-computer interface (BCI) application. Successful application of CSP depends on the filter band selection to a large degree. However, the most proper band is typically subject-specific and can hardly be determined manually.

NEW METHOD

This study proposes a sparse filter band common spatial pattern (SFBCSP) for optimizing the spatial patterns. SFBCSP estimates CSP features on multiple signals that are filtered from raw EEG data at a set of overlapping bands. The filter bands that result in significant CSP features are then selected in a supervised way by exploiting sparse regression. A support vector machine (SVM) is implemented on the selected features for MI classification.

RESULTS

Two public EEG datasets (BCI Competition III dataset IVa and BCI Competition IV IIb) are used to validate the proposed SFBCSP method. Experimental results demonstrate that SFBCSP help improve the classification performance of MI.

COMPARISON WITH EXISTING METHODS

The optimized spatial patterns by SFBCSP give overall better MI classification accuracy in comparison with several competing methods.

CONCLUSIONS

The proposed SFBCSP is a potential method for improving the performance of MI-based BCI.

Probabilistic Common Spatial Patterns for Multichannel EEG Analysis

Common spatial patterns (CSP) is a well-known spatial filtering algorithm for multichannel electroencephalogram (EEG) analysis. In this paper, we cast the CSP algorithm in a probabilistic modeling setting. Specifically, probabilistic CSP (P-CSP) is proposed as a generic EEG spatio-temporal modeling framework that subsumes the CSP and regularized CSP algorithms. The proposed framework enables us to resolve the overfitting issue of CSP in a principled manner. We derive statistical inference algorithms that can alleviate the issue of local optima. In particular, an efficient algorithm based on eigendecomposition is developed for maximum a posteriori (MAP) estimation in the case of isotropic noise. For more general cases, a variational algorithm is developed for group-wise sparse Bayesian learning for the P-CSP model and for automatically determining the model size. The two proposed algorithms are validated on a simulated data set. Their practical efficacy is also demonstrated by successful applications to single-trial classifications of three motor imagery EEG data sets and by the spatio-temporal pattern analysis of one EEG data set recorded in a Stroop color naming task.

Single-trial classification of event-related potentials in rapid serial visual presentation tasks using supervised spatial filtering

Accurate detection of single-trial event-related potentials (ERPs) in the electroencephalogram (EEG) is a difficult problem that requires efficient signal processing and machine learning techniques. Supervised spatial filtering methods that enhance the discriminative information in EEG data are commonly used to improve single-trial ERP detection. We propose a convolutional neural network (CNN) with a layer dedicated to spatial filtering for the detection of ERPs and with training based on the maximization of the area under the receiver operating characteristic curve (AUC). The CNN is compared with three common classifiers: 1) Bayesian linear discriminant analysis; 2) multilayer perceptron (MLP); and 3) support vector machines. Prior to classification, the data were spatially filtered with xDAWN (for the maximization of the signal-to-signal-plus-noise ratio), common spatial pattern, or not spatially filtered. The 12 analytical techniques were tested on EEG data recorded in three rapid serial visual presentation experiments that required the observer to discriminate rare target stimuli from frequent nontarget stimuli. Classification performance discriminating targets from nontargets depended on both the spatial filtering method and the classifier. In addition, the nonlinear classifier MLP outperformed the linear methods. Finally, training based AUC maximization provided better performance than training based on the minimization of the mean square error. The results support the conclusion that the choice of the systems architecture is critical and both spatial filtering and classification must be considered together.

Adaptation of motor imagery EEG classification model based on tensor decomposition

OBJECTIVE

Session-to-session nonstationarity is inherent in brain-computer interfaces based on electroencephalography. The objective of this paper is to quantify the mismatch between the training model and test data caused by nonstationarity and to adapt the model towards minimizing the mismatch.

APPROACH

We employ a tensor model to estimate the mismatch in a semi-supervised manner, and the estimate is regularized in the discriminative objective function.

MAIN RESULTS

The performance of the proposed adaptation method was evaluated on a dataset recorded from 16 subjects performing motor imagery tasks on different days. The classification results validated the advantage of the proposed method in comparison with other regularization-based or spatial filter adaptation approaches. Experimental results also showed that there is a significant correlation between the quantified mismatch and the classification accuracy.

SIGNIFICANCE

The proposed method approached the nonstationarity issue from the perspective of data-model mismatch, which is more direct than data variation measurement. The results also demonstrated that the proposed method is effective in enhancing the performance of the feature extraction model.