Independent Vector Analysis for Feature Extraction in Motor Imagery Classification.

Electroencephalogram (EEG) based motor imagery (MI) brain-computer interface (BCI) has emerged as a promising tool for communication and control. Most MI classification methods use fixed-length time windows to intercept signals and perform subsequent analyses. However, the fixed-length time window interception method can not achieve optimal performance due to significant differences in the multiple imagining tasks of the same subject. In this paper, we present a novel interception method using long and short windows (LSWs). This method takes advantage of the subject’s motor imaginary strength at different times of the task to select specific time windows corresponding to the most salient features. The features corresponding to the selected time windows are used for the final MI classification. We compare the proposed LSW interception method with the fixed-length time window method on a public EEG dataset (BCI competition IV dataset 1) and a self-collected dataset. The results show that the classification accuracies are improved with the LSW interception method on both datasets. When using the support vector machine (SVM) classifier, the classification accuracy of common spatial pattern with the LSW method achieves 2.57% and 1.12% improvement on two datasets, respectively, and the classification accuracy of filter bank common spatial pattern (FBCSP) with the LSW method achieves 0.93% and 1.48% improvement, respectively. Among them, the classification accuracy of the LSW method with FBCSP and SVM is the highest, which is 93.43% and 91.12%, respectively. Compared with the traditional methods, this method significantly increases the classification accuracy and provides a new idea for researching the MI classification method in BCI.

In recent years, Brain Computer Interface (BCI) based on motor imagery has been widely used in the fields of medicine, active safe systems for automobiles, entertainment, and so on. Motor imagery relevant electroencephalogram (EEG) signals are weak, nonlinear, and susceptible to interference. As a feature extraction method for motor imagery, Common Spatial Pattern (CSP) has been proven to be very effective. However, its effectiveness depends heavily on the choice of frequency bands, and Euclidean space cannot effectively describe the inner relationship. To solve these problems, a classification approach for motor imagery using the Geodesic Filtering Common Spatial Pattern (GFCSP) and filter-bank Feature Weighted Support Vector Machine (FWSVM) is presented. First, GFCSP based on the Riemannian manifold is proposed, in which the extracted covariance features are spatially filtered in Riemannian tangent space, and the average covariance matrix is replaced by Riemannian mean in CSP. Second, filter-bank FWSVM with a feature weighted matrix is proposed. EEG signals are filtered into 8-12 Hz, 12-16 Hz, 18-22 Hz, 22-26 Hz, and a wide band of 8-24 Hz, and GFCSP features of these filtered signals are extracted. A feature weighted matrix is calculated using mutual information and the Pearson correlation coefficient from these features and class information. Then, the Support Vector Machine (SVM) is used for classification with the feature weighted matrix. Finally, the proposed method is validated on the dataset IVa in BCI competition III. Classification accuracies of the five subjects are 92.31%, 99.03%, 80.36%, 96.30%, and 97.67%, which demonstrate the effectiveness of our proposed method.

Motor Imagery (MI) based Brain Computer Inter-face (BCI) is a promising neurorehabilitation tool for treating motor impaired stroke survivors. It enables the MI electroencephalogram (EEG) signals to be converted/mapped into customized robotic and assisting commands. Even though stroke causes varying effects in the brain, the EEG signals have shown promises towards the MI classification of post-stroke subjects. This paper presents a MI based left and right wrist dorsiflexion classification performed on 6 stoke subjects. The MI EEG data are recorded from 16 electrode locations on the subject’s scalp. Three different feature extraction methods are used to compare and find the best performing one; the first one is based on the Wavelet Packet Decomposition (WPD) combined with Higher Order Statistics (HOS) which is compared with the widely used Common Spatial Pattern (CSP) and Filter Bank Common Spatial Pattern (FBCSP) filter method. The MI classifications are performed using Random Forest (RF) algorithm to achieve a mean accuracy that exceeds 70% for the WPD+HOS method, while outperforming the CSP and FBCSP based methods.

Effective common spatial pattern (CSP) feature extraction for motor-imagery (MI) EEG recordings usually depends on the filter band selection to a large extent. However, the most proper filter band can hardly be determined manually due to its subject-specific property. This study introduces a sparse support vector machine (SSVM) approach to implement simultaneous CSP feature selection and classification for MI-based brain–computer interface (BCI). In SSVM, CSP features are first extracted on multiple signals that are filtered from raw EEG data at a set of overlapping subbands. SVM with l 1-norm regularization is then proposed to classify MI tasks with automatic selection of filter bands giving the significant CSP features. The effectiveness of SSVM for MI classification is demonstrated on the BCI Competition III dataset IVa, in comparison with several other competing methods. Experimental results indicate that the proposed SSVM method is promising for development of an improved MI-based BCI.

Independent vector analysis (IVA) is a special form of independent component analysis (ICA), which has demonstrated its prominent performance in solving convolutive blind source separation (BSS) problems in the frequency domain. Most IVA algorithms are based on optimizing certain contrast functions, where the main difficulty of these approaches lies in finding a reliable and fast estimation of the unknown distribution of sources. Despite the rich availability of efficient tensorial approaches to the standard ICA problem, these methods have not been explored considerably for IVA. In this article, we propose a matrix joint diagonalization approach to solve the complex IVA problem. The new factorization neither relies on a whitening process, nor does it require an estimate of the joint probability distribution of the dependent signal groups. The latter is in contrast to most IVA approaches up to date. The underlying geometry of the problem is investigated together with a critical point analysis of the resulting cost function. A conjugate gradient algorithm on the appropriate manifold setting is developed.

Graph vertex and spectral features for EEG-based motor imagery classification.

Event-related desynchronization/synchronization patterns during right/left motor imagery (MI) are effective features for an electroencephalogram-based brain-computer interface (BCI). As MI tasks are subject-specific, selection of subject-specific discriminative frequency components play a vital role in distinguishing these patterns. This paper proposes a new discriminative filter bank (FB) common spatial pattern algorithm to extract subject-specific FB for MI classification. The proposed method enhances the classification accuracy in BCI competition III dataset IVa and competition IV dataset IIb. Compared to the performance offered by the existing FB-based method, the proposed algorithm offers error rate reductions of 17.42% and 8.9% for BCI competition datasets III and IV, respectively.

Independent vector analysis followed by HMM-based feature enhancement for robust speech recognition

Motor imagery (MI) is the most used paradigm in Brain-Computer Interface (BCI). As in other BCI paradigms, it is divided into feature extraction and classification. In this study, we implemented two feature extraction methods, one based on the well-known Common Spatial Pattern (CSP) algorithm and the other based on a not so well-known electroencephalogram (EEG) source localization technique called Low-Resolution Brain Electromagnetic Tomography (LORETA). We evaluated 30 right-handed subjects from an EEG data set made publicly available through Giga Science, where participants performed MI of left- and right-hand movements. After feature extraction with CSP and LORETA, MI tasks classification were carried out using the Linear Discriminant Analysis (LDA), Support Vector Machines (SVM), Naïve Bayes (NB) and Multi-Layer Perceptron (MLP) algorithms. Finally, we evaluated classification performance with all possible combination of classifiers and feature extraction methods. For all classifiers, the CSP feature extraction method performed better than LORETA. The best classification accuracy for the LORETA method was 71.2% and for the CSP method was 94.2%, both achieved with SVM.

This paper is focused on the experimental approach to explore the potential of independent component analysis (ICA) in the context of motor imagery (MI)-based brain-computer interface (BCI). We presented a simple and efficient algorithmic framework of ICA-based MI BCI (ICA-MIBCI) for the evaluation of four classical ICA algorithms (Infomax, FastICA, Jade, and Sobi) as well as a simplified Infomax (sInfomax). Two novel performance indexes, self-test accuracy and the number of invalid ICA filters, were employed to assess the performance of MIBCI based on different ICA variants. As a reference method, common spatial pattern (CSP), a commonly-used spatial filtering method, was employed for the comparative study between ICA-MIBCI and CSP-MIBCI. The experimental results showed that sInfomax-based spatial filters exhibited significantly better transferability in session to session and subject to subject transfer as compared to CSP-based spatial filters. The online experiment was also introduced to demonstrate the practicability and feasibility of sInfomax-based MIBCI. However, four classical ICA variants, especially FastICA, Jade, and Sobi, performed much worse as compared to sInfomax and CSP in terms of classification accuracy and stability. We consider that conventional ICA-based spatial filtering methods tend to be overfitting while applied to real-life electroencephalogram data. Nevertheless, the sInfomax-based experimental results indicate that ICA methods have a great space for improvement in the application of MIBCI. We believe that this paper could bring forth new ideas for the practical implementation of ICA-MIBCI.

One of the essential challenges in brain–computer interface is to classify motor imagery (MI) signals. In this paper, an ensemble SVM-based voting system is proposed. In each line of this system, the EEG signal is transformed into different representations based on discrete cosine transform, Fourier transform, common spatial pattern, and empirical mode decomposition, and then these representations are combined in a triple-frame matrix. These frames are fed into a pre-trained deep convolutional neural network as a feature extractor. For each line, an SVM is employed to classify the extracted feature vectors. Finally, a decision is made based on voting between these SVMs. Performance of the proposed method is examined on the BCI Competition III dataset Iva to separate right hand and foot movement imagery. The simple proposed method achieves the average accuracy of 96.34% for all of the subjects, and 99.70% for the best situation that is an improvement in MI classification. In addition, it can be seen that right side of the brain is more effective than the other side in EEG-based MI classification.

Analysis of time-evolving data is crucial to understand the functioning of dynamic systems such as the brain. For instance, analysis of functional magnetic resonance imaging (fMRI) data collected during a task may reveal spatial regions of interest, and how they evolve during the task. However, capturing underlying spatial patterns as well as their change in time is challenging. The traditional approach in fMRI data analysis is to assume that underlying spatial regions of interest are static. In this article, using fractional amplitude of low-frequency fluctuations (fALFF) as an effective way to summarize the variability in fMRI data collected during a task, we arrange time-evolving fMRI data as a subjects by voxels by time windows tensor, and analyze the tensor using a tensor factorization-based approach called a PARAFAC2 model to reveal spatial dynamics. The PARAFAC2 model jointly analyzes data from multiple time windows revealing subject-mode patterns, evolving spatial regions (also referred to as networks) and temporal patterns. We compare the PARAFAC2 model with matrix factorization-based approaches relying on independent components, namely, joint independent component analysis (ICA) and independent vector analysis (IVA), commonly used in neuroimaging data analysis. We assess the performance of the methods in terms of capturing evolving networks through extensive numerical experiments demonstrating their modeling assumptions. In particular, we show that (i) PARAFAC2 provides a compact representation in all modes, i.e., subjects, time, and voxels, revealing temporal patterns as well as evolving spatial networks, (ii) joint ICA is as effective as PARAFAC2 in terms of revealing evolving networks but does not reveal temporal patterns, (iii) IVA's performance depends on sample size, data distribution and covariance structure of underlying networks. When these assumptions are satisfied, IVA is as accurate as the other methods, (iv) when subject-mode patterns differ from one time window to another, IVA is the most accurate. Furthermore, we analyze real fMRI data collected during a sensory motor task, and demonstrate that a component indicating statistically significant group difference between patients with schizophrenia and healthy controls is captured, which includes primary and secondary motor regions, cerebellum, and temporal lobe, revealing a meaningful spatial map and its temporal change.

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.

Recognition of motor imagery tasks for BCI using CSP and chaotic PSO twin SVM

In motor imagery (MI) based brain-computer interface (BCI), common spatial pattern (CSP) is most popularly used for discriminant feature extraction. However, the performance of CSP depends on the operational frequency bands, which are selected manually or set to a broad frequency range in most of the previously developed applications. Due to subject to subject or even trial to trial variability of frequency band affected by MI task, these methods suffer from poor performance. We have proposed a novel approach, using combination of multivariate iterative filtering (MIF) and CSP (MIFCSP), to automatically select optimal frequency bands based on MIF which can be further used for discriminant feature extraction. MIF decomposes the signal into several multivariate intrinsic mode functions, from which features are extracted using CSP. We select the minimum number of most significant features for which highest classification accuracy is achieved. Subsequently, linear discriminant analysis (LDA) classifier is used to classify different MI tasks. Experimental results for BCI competition IV dataset 2a and BCI competition III-IIIa are presented. For left hand versus right hand MI classification, proposed MIFCSP method provides 83.18% and 84.44% average accuracy, respectively. Superior classification performance confirms that MIFCSP is a promising candidate for MI BCI application.