A novel hybrid kernel function relevance vector machine for multi-task motor imagery EEG classification

Hybrid optimization assisted channel selection of EEG for deep learning model-based classification of motor imagery task

OBJECTIVES

To design and develop an approach named HC + SMA-SSA scheme for classifying motor imagery task.

METHODS

The offered model employs a new method for classifying motor imagery task. Initially, down sampling is deployed to pre-process the incoming signal. Subsequently, "Modified Stockwell Transform (ST) and common spatial pattern (CSP) based features are extracted". Then, optimal channel selection is made by a novel hybrid optimization model named as Spider Monkey Assisted SSA (SMA-SSA). Here, "Long Short Term Memory (LSTM) and Bidirectional Gated Recurrent Unit (BI-GRU)" models are used for final classification, whose outcomes are averaged at the end. At last, the improvement of SMA-SSA based model is proven over different metrics.

RESULTS

A superior sensitivity of 0.939 is noted for HC + SMA-SSA that was higher over HC with no optimization and proposed with traditional ST.

CONCLUSIONS

The proposed method achieved effective classification performance in terms of performance measures.

Raman spectroscopy-based prediction of ofloxacin concentration in solution using a novel loss function and an improved GA-CNN model

BACKGROUND

A Raman spectroscopy method can quickly and accurately measure the concentration of ofloxacin in solution. This method has the advantages of accuracy and rapidity over traditional detection methods. However, the manual analysis methods for the collected Raman spectral data often ignore the nonlinear characteristics of the data and cannot accurately predict the concentration of the target sample.

METHODS

To address this drawback, this paper proposes a novel kernel-Huber loss function that combines the Huber loss function with the Gaussian kernel function. This function is used with an improved genetic algorithm-convolutional neural network (GA-CNN) to model and predict the Raman spectral data of different concentrations of ofloxacin in solution. In addition, the paper introduces recurrent neural networks (RNN), long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM) and gated recurrent units (GRU) models to conduct multiple experiments and use root mean square error (RMSE) and residual predictive deviation (RPD) as evaluation metrics.

RESULTS

The proposed method achieved an [Formula: see text] of 0.9989 on the test set data and improved by 3% over the traditional CNN. Multiple experiments were also conducted using RNN, LSTM, BiLSTM, and GRU models and evaluated their performance using RMSE, RPD, and other metrics. The results showed that the proposed method consistently outperformed these models.

CONCLUSIONS

This paper demonstrates the effectiveness of the proposed method for predicting the concentration of ofloxacin in solution based on Raman spectral data, in addition to discussing the advantages and limitations of the proposed method, and the study proposes a solution to the problem of deep learning methods for Raman spectral concentration prediction.

3D Convolution neural network with multiscale spatial and temporal cues for motor imagery EEG classification

Recently, deep learning-based methods have achieved meaningful results in the Motor imagery electroencephalogram (MI EEG) classification. However, because of the low signal-to-noise ratio and the various characteristics of brain activities among subjects, these methods lack a subject adaptive feature extraction mechanism. Another issue is that they neglect important spatial topological information and the global temporal variation trend of MI EEG signals. These issues limit the classification accuracy. Here, we propose an end-to-end 3D CNN to extract multiscale spatial and temporal dependent features for improving the accuracy performance of 4-class MI EEG classification. The proposed method adaptively assigns higher weights to motor-related spatial channels and temporal sampling cues than the motor-unrelated ones across all brain regions, which can prevent influences caused by biological and environmental artifacts. Experimental evaluation reveals that the proposed method achieved an average classification accuracy of 93.06% and 97.05% on two commonly used datasets, demonstrating excellent performance and robustness for different subjects compared to other state-of-the-art methods.In order to verify the real-time performance in actual applications, the proposed method is applied to control the robot based on MI EEG signals. The proposed approach effectively addresses the issues of existing methods, improves the classification accuracy and the performance of BCI system, and has great application prospects.

Application of multi-task transfer learning: The combination of EA and optimized subband regularized CSP to classification of 8-channel EEG signals with small dataset

Introduction

The volume conduction effect and high dimensional characteristics triggered by the excessive number of channels of EEG cap-acquired signals in BCI systems can increase the difficulty of classifying EEG signals and the lead time of signal acquisition. We aim to combine transfer learning to decode EEG signals in the few-channel case, improve the classification performance of the motor imagery BCI system across subject cases, reduce the cost of signal acquisition performed by the BCI system, and improve the usefulness of the system.

Methods

Dataset2a from BCI CompetitionIV(2008) was used as Dataset1, and our team's self-collected dataset was used as Dataset2. Dataset1 acquired EEG signals from 9 subjects using a 22-channel device with a sampling frequency of 250 Hz. Dataset2 acquired EEG signals from 10 healthy subjects (8 males and 2 females; age distribution between 21-30 years old; mean age 25 years old) using an 8-channel system with a sampling frequency of 1000 Hz. We introduced EA in the data preprocessing process to reduce the signal differences between subjects and proposed VFB-RCSP in combination with RCSP and FBCSP to optimize the effect of feature extraction.

Results

Experiments were conducted on Dataset1 with EEG data containing only 8 channels and achieved an accuracy of 78.01 and a kappa coefficient of 0.54. The accuracy exceeded most of the other methods proposed in recent years, even though the number of channels used was significantly reduced. On Dataset 2, an accuracy of 59.77 and a Kappa coefficient of 0.34 were achieved, which is a significant improvement compared to other poorly improved classical protocols.

Discussion

Our work effectively improves the classification of few-channel EEG data. It overcomes the dependence of existing algorithms on the number of channels, the number of samples, and the frequency band, which is significant for reducing the complexity of BCI models and improving the user-friendliness of BCI systems.

A multi-modal brain-computer interface based on threshold discrimination and its application in wheelchair control

In this study, we propose a novel multi-modal brain-computer interface (BCI) system based on the threshold discrimination, which is proposed for the first time to distinguish between SSVEP and MI potentials. The system combines these two heterogeneous signals to increase the number of control commands and improve the performance of asynchronous control of external devices. In this research, an electric wheelchair is controlled as an example. The user can continuously control the wheelchair to turn left/right through motion imagination (MI) by imagining left/right-hand movement and generate another 6 commands for the wheelchair control by focusing on the SSVEP stimulation panel. Ten subjects participated in a MI training session and eight of them completed a mobile obstacle-avoidance experiment in a complex environment requesting high control accuracy for successful manipulation. Comparing with the single-modal BCI-controlled wheelchair system, the results demonstrate that the proposed multi-modal method is effective by providing more satisfactory control accuracy, and show the potential of BCI-controlled systems to be applied in complex daily tasks.

Spatial and Temporal Evolution Analysis of Industrial Green Technology Innovation Efficiency in the Yangtze River Economic Belt

As a fusion point of innovation-driven green development, green technology innovation has become an essential engine for green transformation and high-quality economic development of the Yangtze River Economic Belt. Based on the panel data of 110 cities in the Yangtze River Economic Belt from 2006 to 2020, this paper uses the super-SBM model to measure the efficiency of industrial green technology innovation. Then, the Dagum Gini coefficient and its subgroup decomposition method, kernel density estimation, and the spatial Markov chain will discuss the convergence characteristics and dynamic evolution law of industrial green technology innovation efficiency in the Yangtze River Economic Belt. The results indicate several key points. (1) On the whole, the industrial green innovation efficiency of the Yangtze River Economic Belt shows a trend of the “N” type, which increases slowly at first and then decreases and then increases, and shows a non-equilibrium feature of “east high and west low” in space. (2) The average GML index of industrial green technology innovation efficiency in the Yangtze River Economic Belt is greater than 1, and technological progress is the main driving force in promoting efficiency growth. (3) There are spatial and temporal differences in industrial green technological innovation efficiency in the Yangtze River Economic Belt. Interregional differences and hypervariable density are the primary sources of overall differences. (4) During the study period, the absolute difference in industrial green technology innovation efficiency among regions showed a trend of “expansion-reduction-expansion”, and the innovation efficiency gradually converged to a single equilibrium point. (5) The industrial green technology innovation efficiency transfer in the Yangtze River Economic Belt shows a specific spatial dependence. Accordingly, policy suggestions are put forward to further improve industrial green technological innovation in the Yangtze River Economic Belt.

Tetromino pattern based accurate EEG emotion classification model

Nowadays, emotion recognition using electroencephalogram (EEG) signals is becoming a hot research topic. The aim of this paper is to classify emotions of EEG signals using a novel game-based feature generation function with high accuracy. Hence, a multileveled handcrafted feature generation automated emotion classification model using EEG signals is presented. A novel textural features generation method inspired by the Tetris game called Tetromino is proposed in this work. The Tetris game is one of the famous games worldwide, which uses various characters in the game. First, the EEG signals are subjected to discrete wavelet transform (DWT) to create various decomposition levels. Then, novel features are generated from the decomposed DWT sub-bands using the Tetromino method. Next, the maximum relevance minimum redundancy (mRMR) features selection method is utilized to select the most discriminative features, and the selected features are classified using support vector machine classifier. Finally, each channel's results (validation predictions) are obtained, and the mode function-based voting method is used to obtain the general results. We have validated our developed model using three databases (DREAMER, GAMEEMO, and DEAP). We have attained 100% accuracies using DREAMER and GAMEEMO datasets. Furthermore, over 99% of classification accuracy is achieved for DEAP dataset. Thus, our developed emotion detection model has yielded the best classification accuracy rate compared to the state-of-the-art techniques and is ready to be tested for clinical application after validating with more diverse datasets. Our results show the success of the presented Tetromino pattern-based EEG signal classification model validated using three public emotional EEG datasets.

A novel method to reduce the motor imagery BCI illiteracy

To reduce the motor imagery brain-computer interface (MI-BCI) illiteracy phenomenon and improve the classification accuracy, this paper proposed a novel method combining paradigm selection and Riemann distance classification. Firstly, a novel sensitivity-based paradigm selection (SPS) algorithm is designed for the optimization of classification to find the best classification pattern through a sensitive indicator. Then, a generalized Riemann minimum distance mean (GRMDM) classifier is proposed by introducing a weight factor to fuse the Log-Euclidean Metric classifier and the Riemannian Stein divergence classifier. The experimental results show that the proposed method achieves a better performance for multi-class motor imagery tasks. The average classification accuracy on the BCI competition IV dataset2a is 80.98%, which is 11.04% higher than Stein divergence classifier on the original two-class paradigm. Furthermore, the proposed method demonstrates its capacity on reducing MI-BCI illiteracy. Graphical abstract Here we investigate whether the BCI illiteracy phenomenon can be reduced through sensitivity-based paradigm selection (SPS) method and generalized Riemann minimum distance mean (GRMDM) classifier when performing motor imagery tasks.

Multiscale space-time-frequency feature-guided multitask learning CNN for motor imagery EEG classification

OBJECTIVE

Motor imagery (MI) electroencephalography (EEG) classification is regarded as a promising technology for brain--computer interface (BCI) systems, which help people to communicate with the outside world using neural activities. However, decoding human intent accurately is a challenging task because of its small signal-to-noise ratio and non-stationary characteristics. Methods that directly extract features from raw EEG signals ignores key frequency domain information. One of the challenges in MI classification tasks is finding a way to supplement the frequency domain information ignored by the raw EEG signal.

APPROACH

In this study, we fuse different models using their complementary characteristics to develop a multiscale space-time-frequency feature-guided multitask learning convolutional neural network (CNN) architecture. The proposed method consists of four modules: the space-time feature-based representation module, time-frequency feature-based representation module, multimodal fused feature-guided generation module, and classification module. The proposed framework is based on multitask learning. The four modules are trained using three tasks simultaneously and jointly optimised.

RESULTS

The proposed method is evaluated using three public challenge datasets. Through quantitative analysis, we demonstrate that our proposed method outperforms most state-of-the-art machine learning and deep learning techniques for EEG classification, thereby demonstrating the robustness and effectiveness of our method. Moreover, the proposed method is employed to realize control of robot based on EEG signal, verifying its feasibility in real-time applications.

SIGNIFICANCE

To the best of our knowledge, a deep CNN architecture that fuses different input cases, which have complementary characteristics, has not been applied to BCI tasks. Because of the interaction of the three tasks in the multitask learning architecture, our method can improve the generalisation and accuracy of subject-dependent and subject-independent methods with limited annotated data.

The CSP-Based New Features Plus Non-Convex Log Sparse Feature Selection for Motor Imagery EEG Classification

The common spatial pattern (CSP) is a very effective feature extraction method in motor imagery based brain computer interface (BCI), but its performance depends on the selection of the optimal frequency band. Although a lot of research works have been proposed to improve CSP, most of these works have the problems of large computation costs and long feature extraction time. To this end, three new feature extraction methods based on CSP and a new feature selection method based on non-convex log regularization are proposed in this paper. Firstly, EEG signals are spatially filtered by CSP, and then three new feature extraction methods are proposed. We called them CSP-wavelet, CSP-WPD and CSP-FB, respectively. For CSP-Wavelet and CSP-WPD, the discrete wavelet transform (DWT) or wavelet packet decomposition (WPD) is used to decompose the spatially filtered signals, and then the energy and standard deviation of the wavelet coefficients are extracted as features. For CSP-FB, the spatially filtered signals are filtered into multiple bands by a filter bank (FB), and then the logarithm of variances of each band are extracted as features. Secondly, a sparse optimization method regularized with a non-convex log function is proposed for the feature selection, which we called LOG, and an optimization algorithm for LOG is given. Finally, ensemble learning is used for secondary feature selection and classification model construction. Combing feature extraction and feature selection methods, a total of three new EEG decoding methods are obtained, namely CSP-Wavelet+LOG, CSP-WPD+LOG, and CSP-FB+LOG. Four public motor imagery datasets are used to verify the performance of the proposed methods. Compared to existing methods, the proposed methods achieved the highest average classification accuracy of 88.86, 83.40, 81.53, and 80.83 in datasets 1–4, respectively. The feature extraction time of CSP-FB is the shortest. The experimental results show that the proposed methods can effectively improve the classification accuracy and reduce the feature extraction time. With comprehensive consideration of classification accuracy and feature extraction time, CSP-FB+LOG has the best performance and can be used for the real-time BCI system.