Filter bank CSP with Riemannian weighting for disability-centric motor imagery brain computer interface.

Comparative analysis of spectral and temporal combinations in CSP-based methods for decoding hand motor imagery tasks

Traditional common spatial pattern (CSP) algorithms for upper limb neural rehabilitation face inherent challenges of overlapping cortical representations and frequency sensitivity, which hinder the decoding performance of motor imagery (MI) electroencephalogram (EEG) signals. To address these issues, this study adopts an improved discriminative filter bank CSP (DFBCSP) framework and applies it to the decoding of upper limb MI-EEG signals, achieving remarkable classification performance. EEG data were acquired from sixteen participants performing two-class (left upper limb flexion-extension vs. relaxing) and three-class (left upper limb flexion vs. right upper limb extension vs. relaxing) MI tasks. The acquired EEG data were then decomposed into nine distinct sub-bands, followed by the adoption of a mutual information-based feature selection strategy to optimize the feature sets. These optimized feature sets were subsequently input into three classification models, namely multilayer perceptron (MLP), support vector machine (SVM), and linear discriminant analysis (LDA), for MI task classification. Experimental results demonstrate that the DFBCSP + MLP method significantly outperforms the traditional CSP approach. Specifically, it achieves an accuracy of 94.83% (Kappa coefficient: 0.890) in two-class MI tasks and 86.20% (Kappa coefficient: 0.775) in three-class MI tasks. The DFBCSP + MLP framework exhibits high robustness and provides a potential technical framework and theoretical basis for future research on the rehabilitation of patients with upper limb motor dysfunction.

BackgroundAs a typical self-paced brain–computer interface (BCI) system, the motor imagery (MI) BCI has been widely applied in fields such as robot control, stroke rehabilitation, and assistance for patients with stroke or spinal cord injury. Many studies have focused on the traditional spatial filters obtained through the common spatial pattern (CSP) method. However, the CSP method can only obtain fixed spatial filters for specific input signals. In addition, the CSP method only focuses on the variance difference of two types of electroencephalogram (EEG) signals, so the decoding ability of EEG signals is limited.MethodsTo make up for these deficiencies, this study introduces a novel spatial filter-solving paradigm named adaptive spatial pattern (ASP), which aims to minimize the energy intra-class matrix and maximize the inter-class matrix of MI-EEG after spatial filtering. The filter bank adaptive and common spatial pattern (FBACSP), our proposed method for MI-EEG decoding, amalgamates ASP spatial filters with CSP features across multiple frequency bands. Through a dual-stage feature selection strategy, it employs the Particle Swarm Optimization algorithm for spatial filter optimization, surpassing traditional CSP approaches in MI classification. To streamline feature sets and enhance recognition efficiency, it first prunes CSP features in each frequency band using mutual information, followed by merging these with ASP features.ResultsComparative experiments are conducted on two public datasets (2a and 2b) from BCI competition IV, which show the outstanding average recognition accuracy of FBACSP. The classification accuracy of the proposed method has reached 74.61 and 81.19% on datasets 2a and 2b, respectively. Compared with the baseline algorithm, filter bank common spatial pattern (FBCSP), the proposed algorithm improves by 11.44 and 7.11% on two datasets, respectively (p < 0.05).ConclusionIt is demonstrated that FBACSP has a strong ability to decode MI-EEG. In addition, the analysis based on mutual information, t-SNE, and Shapley values further proves that ASP features have excellent decoding ability for MI-EEG signals and explains the improvement of classification performance by the introduction of ASP features. These findings may provide useful information to optimize EEG-based BCI systems and further improve the performance of non-invasive BCI.

In this paper, the four-class classification (Left hand, right hand, feet, and tongue) Motor Imagery (MI) signals is performed using four different feature extraction techniques. First, raw EEG signals are pre-processed using the Multi-Class Common Spatial Pattern (CSP) method (one-versus-rest scheme), which discriminates features in feature space and improves the accuracy of classification. Then, four different features, namely channel FFT energy, mean band power, mean channel energy, and Discrete Wavelet Transform (DWT) based mean band energy features, are extracted from pre-processed EEG signals and compared to find the most suitable feature for the discrimination of four-class MI tasks. Besides, three classifiers, namely Bayesian Network classifier (Naive Bays), Linear discriminant analysis (LDA), and Linear Support Vector Machine (SVM), are compared. Performance evaluation is done on BCI competition IV dataset 2a using classification accuracy along with different performance measures calculated from a confusion matrix. The performance of the LDA classifier is found better than linear SVM and Naive Bays. The presented framework with DWT as a feature extraction technique and LDA classifier has obtained an average test classification accuracy of 80.29% over four subjects out of nine. The less computational cost of this framework makes it suitable for online Motor Imagery based BCI Systems.KeywordsElectroencephalogram signals (EEG)Motor imagery (MI)Brain computer interface (BCI)Discrete wavelet transform (DWT)Common spatial pattern (CSP)

Sub-band target alignment common spatial pattern in brain-computer interface

Motor imagery based brain computer interface (BCI) has drawback of long subject dependent calibration session times. This can be a very exhausting and a time consuming process. In order to alleviate it, transfer learning and active learning approaches can be utilised. Informative instances are selected by applying active learning concept from other subjects under similar circumstances. Then, they are transferred to target user domain which has low number of training data. This informative transfer learning approach is associated with common spatial pattern (CSP) as feature extraction method in our previous attempt. CSP features are widely used for motor imagery-based BCI systems. However, the classical CSP algorithm will perform poorly when operational frequency bands are inadequately selected. Therefore, in the present study, filter bank common spatial pattern (FBCSP) algorithm has been applied for extracting features from the multi-class motor imagery data. FBCSP algorithm selects subject-specific operational frequency bands for extracting discriminative features. We incorporated FBCSP features into informative instance transfer learning framework to investigate the effect of subject specific feature selection. Results show that performance of new users can be improved with reduced number of training samples when FBCSP features are used compared to the classical CSP-based features.

Deep learning technology is rapidly spreading in recent years and has been extensive attempts in the field of Brain-Computer Interface (BCI). Though the accuracy of Motor Imagery (MI) BCI systems based on the deep learning have been greatly improved compared with some traditional algorithms, it is still a big problem to clearly interpret the deep learning models. To address the issues, this work first introduces a popular deep learning model EEGNet and compares it with the traditional algorithm Filter-Bank Common Spatial Pattern (FBCSP). After that, this work considers that the 1-D convolution of EEGNet can be explained by a special Discrete Wavelet Transform (DWT), and the depthwise convolution of EEGNet is similar to the Common Spatial Pattern (CSP) algorithm. Therefore, this work improves the EEGNet by using the algorithm Temporary Constrained Sparse Group Lasso (TCSGL) to enhance its performance. The proposed model TSGL-EEGNet is tested on the BCI Competition IV 2a and BCI Competition III IIIa datasets that both are 4-classes classification MI tasks. The testing results show that the proposed model has achieved 78.96% (0.7194) average classification accuracy (kappa) on the dataset BCI Competition IV 2a, which are greater than EEGNet, C2CM, MB3DCNN, SS-MEMDBF and FBCSP, especially on insensitive subjects. The proposed model has also achieved 85.30% (0.8040) average classification accuracy (kappa) on the dataset BCI Competition III IIIa, which are greater than the EEGNet, MFTFS et al. At last, this work uses average-validation and stacking to further enhance the effect of the model. The 4-classes classification average accuracy rates reach 81.34% and 88.89%, and the kappas reach 0.7511 and 0.8519 on dataset BCI Competition IV 2a and BCI Competition III IIIa, respectively. Additionally, this work also uses the Grad-CAM to visualize the frequency and spatial features that are learned by the neural network.

Compared to other electroencephalogram (EEG) modalities, motor imagery (MI) based brain-computer interfaces (BCls) can provide more natural and intuitive communication between human intentions and external machines. However, this type of BCI depends heavily on effective signal processing to discriminate EEG patterns corresponding to various MI tasks, especially feature extraction procedures. In this study, a comparison of different feature extraction methods was conducted for EEG classification of imaginary movements within the same upper extremity. Unlike traditional MI tasks (left/right hand), six imaginary movements from the same unilateral upper extremity were proposed and evaluated, including elbow extension/flexion, wrist pronation/supination, and hand open/grasp. To tackle the classification challenge of MI tasks within the same limb, four types of feature extraction methods were implemented and compared in combination with support vector machine (SVM) and linear discriminant analysis (LDA) classifiers, such as wavelet transformation, power spectrum, autoregressive model, common spatial patterns (CSP) and variants of filter-bank CSP (FBCSP), regularized CSP (RCSP). The overall accuracies of the CSP were significant higher than other three types of feature extraction on a dataset collected from 8 individuals, particularly the SVM with FBCSP had the best performance with an average accuracy of 71.78%. These decoding results of MI tasks during single upper extremity are encouraging and promising in the context of more natural MI-BCI for controlling assisted devices, such as a neuroprosthetic or robotic arm for motor disabled individuals with highly impaired upper extremity.

Signal processing techniques for motor imagery brain computer interface: A review

Background and objectives: Brain-computer interface (BCI) systems typically deploy common spatial pattern (CSP) for feature extraction of mu and beta rhythms based on upper-limbs kinaesthetic motor imageries (KMI). However, it was not used to classify the left versus right foot KMI, due to its location inside the mesial wall of sensorimotor cortex, which makes it difficult to be detected. We report novel classification of mu and beta EEG features, during left and right foot KMI cognitive task, using CSP, and filter bank common spatial pattern (FBCSP) method, to optimize the subject-specific band selection. We initially proposed CSP method, followed by the implementation of FBCSP for optimization of individual spatial patterns, wherein a set of CSP filters was learned, for each of the time/frequency filters in a supervised way. This was followed by the log-variance feature extraction and concatenation of all features (over all chosen spectral-filters). Subsequently, supervised machine learning was implemented, i.e. logistic regression (Logreg) and linear discriminant analysis (LDA), in order to compare the respective foot KMI classification rates. Training and testing data, used in the model, was validated using 10-fold cross validation. Four methodology paradigms are reported, i.e. CSP LDA, CSP Logreg, and FBCSP LDA, FBCSP Logreg. All paradigms resulted in an average classification accuracy rate above the statistical chance level of 60.0% (P < 0.01). On average, FBCSP LDA outperformed remaining paradigms with kappa score of 0.41 and classification accuracy of 70.28% ± 4.23. Similarly, this paradigm enabled discrimination between right and left foot KMI cognitive task at highest accuracy rate i.e. maximum 77.5% with kappa = 0.55 and the area under ROC curve as 0.70 (in single-trial analysis). The proposed novel paradigms, using CSP and FBCSP, established a potential to exploit the left versus right foot imagery classification, in synchronous 2-class BCI for controlling robotic foot, or foot neuroprosthesis.

Common spatial pattern (CSP) is a well-established technique to extract features from electroencephalographic recordings for classification purpose in motor imagery brain-computer interface (BCI).The CSP algorithm is a mathematical procedure used for separating a multivariate signal into additive components which have maximum differences in variance between two windows; in other terms, CSP increases the signal variance for one condition while minimizing the variance for the other condition. Features computed by means of CSP are fed to a data classifier in order to discriminate between two mental tasks. A novel technique to achieve feature extraction is tangentspace mapping (TSM) that insists on spatial covariance matrices computed from the recorded electroencephalogram signals (EEG). TSM is based on Riemannian geometry, which allows one to estimate statistical features of data distributions over non-Euclidean spaces. The aim of this chapter is twofold: first, to provide a new data-visualization tool to visually inspect data distributions on the Riemannian space of spatial covariance matrices and its tangent bundle; second, to present an experimental comparison of CSP and TSM feature extraction, in conjunction with two classification methods, namely, support-vector machine and linear discriminant analysis. In particular, the experimental comparison performed on a number of data sets will show the superiority of TSM-based feature extraction over CSP.

Improvement of brain–computer interface in motor imagery training through the designing of a dynamic experiment and FBCSP

Enhancing motor imagery EEG classification with a Riemannian geometry-based spatial filtering (RSF) method.

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.

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.