scholarly journals Recognition of EEG Signal Motor Imagery Intention Based on Deep Multi-View Feature Learning

Sensors ◽  
2020 ◽  
Vol 20 (12) ◽  
pp. 3496
Author(s):  
Jiacan Xu ◽  
Hao Zheng ◽  
Jianhui Wang ◽  
Donglin Li ◽  
Xiaoke Fang
Keyword(s):  
Frequency Domain ◽  
Motor Imagery ◽  
Recognition Performance ◽  
Feature Learning ◽  
Feature Representation ◽  
Support Vector ◽  
Imagery Task ◽  
Learning Method ◽  
Eeg Signals ◽  
Time Frequency

Recognition of motor imagery intention is one of the hot current research focuses of brain-computer interface (BCI) studies. It can help patients with physical dyskinesia to convey their movement intentions. In recent years, breakthroughs have been made in the research on recognition of motor imagery task using deep learning, but if the important features related to motor imagery are ignored, it may lead to a decline in the recognition performance of the algorithm. This paper proposes a new deep multi-view feature learning method for the classification task of motor imagery electroencephalogram (EEG) signals. In order to obtain more representative motor imagery features in EEG signals, we introduced a multi-view feature representation based on the characteristics of EEG signals and the differences between different features. Different feature extraction methods were used to respectively extract the time domain, frequency domain, time-frequency domain and spatial features of EEG signals, so as to made them cooperate and complement. Then, the deep restricted Boltzmann machine (RBM) network improved by t-distributed stochastic neighbor embedding(t-SNE) was adopted to learn the multi-view features of EEG signals, so that the algorithm removed the feature redundancy while took into account the global characteristics in the multi-view feature sequence, reduced the dimension of the multi-visual features and enhanced the recognizability of the features. Finally, support vector machine (SVM) was chosen to classify deep multi-view features. Applying our proposed method to the BCI competition IV 2a dataset we obtained excellent classification results. The results show that the deep multi-view feature learning method further improved the classification accuracy of motor imagery tasks.

scholarly journals EEG-Based Emotion Recognition Using Quadratic Time-Frequency Distribution

Sensors ◽  
2018 ◽  
Vol 18 (8) ◽  
pp. 2739 ◽  
Author(s):  
Rami Alazrai ◽  
Rasha Homoud ◽  
Hisham Alwanni ◽  
Mohammad Daoud
Keyword(s):  
Emotion Recognition ◽  
Frequency Domain ◽  
Frequency Distribution ◽  
Support Vector ◽  
Eeg Signals ◽  
Time Frequency ◽  
Labeling Schemes ◽  
Frequency Representation ◽  
Frequency Features ◽  
Time Frequency Distribution

Accurate recognition and understating of human emotions is an essential skill that can improve the collaboration between humans and machines. In this vein, electroencephalogram (EEG)-based emotion recognition is considered an active research field with challenging issues regarding the analyses of the nonstationary EEG signals and the extraction of salient features that can be used to achieve accurate emotion recognition. In this paper, an EEG-based emotion recognition approach with a novel time-frequency feature extraction technique is presented. In particular, a quadratic time-frequency distribution (QTFD) is employed to construct a high resolution time-frequency representation of the EEG signals and capture the spectral variations of the EEG signals over time. To reduce the dimensionality of the constructed QTFD-based representation, a set of 13 time- and frequency-domain features is extended to the joint time-frequency-domain and employed to quantify the QTFD-based time-frequency representation of the EEG signals. Moreover, to describe different emotion classes, we have utilized the 2D arousal-valence plane to develop four emotion labeling schemes of the EEG signals, such that each emotion labeling scheme defines a set of emotion classes. The extracted time-frequency features are used to construct a set of subject-specific support vector machine classifiers to classify the EEG signals of each subject into the different emotion classes that are defined using each of the four emotion labeling schemes. The performance of the proposed approach is evaluated using a publicly available EEG dataset, namely the DEAPdataset. Moreover, we design three performance evaluation analyses, namely the channel-based analysis, feature-based analysis and neutral class exclusion analysis, to quantify the effects of utilizing different groups of EEG channels that cover various regions in the brain, reducing the dimensionality of the extracted time-frequency features and excluding the EEG signals that correspond to the neutral class, on the capability of the proposed approach to discriminate between different emotion classes. The results reported in the current study demonstrate the efficacy of the proposed QTFD-based approach in recognizing different emotion classes. In particular, the average classification accuracies obtained in differentiating between the various emotion classes defined using each of the four emotion labeling schemes are within the range of 73 . 8 % – 86 . 2 % . Moreover, the emotion classification accuracies achieved by our proposed approach are higher than the results reported in several existing state-of-the-art EEG-based emotion recognition studies.

Simple Convolutional Neural Network for Left-Right Hands Motor Imagery EEG Signals Classification

2019 ◽  
Vol 13 (3) ◽  
pp. 36-49 ◽  
Author(s):  
Geliang Tian ◽  
Yue Liu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Motor Imagery ◽  
Input Image ◽  
Imagery Task ◽  
Optimal Frequency ◽  
Eeg Signals ◽  
Fisher Discriminant Analysis ◽  
Frequency Bands ◽  
Time Frequency

This article proposes a classification method of two-class motor imagery electroencephalogram (EEG) signals based on convolutional neural network (CNN), in which EEG signals from C3, C4 and Cz electrodes of publicly available BCI competition IV dataset 2b were used to test the performance of the CNN. The authors investigate two similar CNNs: a single-input CNN with a form of 2-dimensional input from short time Fourier transform (STFT) combining time, frequency and location information, and a multiple-input CNN with 3-dimensional input which processes the electrodes as an independent dimension. Fisher discriminant analysis-type F-score based on band pass (BP) feature and power spectra density (PSD) feature are employed respectively to select the subject-optimal frequency bands. In the experiments, typical frequency bands related to motor imagery EEG signals, subject-optimal frequency bands and extension frequency bands are employed respectively as the frequency range of the input image of CNN. The better classification performance of extension frequency bands show that CNN can extract optimal feature from frequency information automatically. The classification result also demonstrates that the proposed approach is more competitive in prediction of left/right hand motor imagery task compared with other state-of-art approaches.

scholarly journals Detection of Epileptic Seizures from EEG Signals by Combining Dimensionality Reduction Algorithms with Machine Learning Models

2021 ◽  
Author(s):  
Muhammad Zubair
Keyword(s):  
Machine Learning ◽  
Dimensionality Reduction ◽  
Frequency Domain ◽  
Work Performance ◽  
Epileptic Seizures ◽  
Dominant Frequency ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Eeg Signals ◽  
Time Frequency

<pre>In this paper, more emphasis has been given to develop a support vector machine (SVM) model using SPPCA and SUBXPCA dimensionality reduction algorithms to increase the classification accuracy. Firstly, Discrete Wavelet Transform (DWT) is applied to EEG signals for extracting the time-frequency domain features of epilepsy such as the energy of each sub-pattern, spike rhythmicity, relative spike amplitude, Dominant Frequency (DF) and Spectral Entropy (SE). The features obtained after performing DWT on an EEG signal are extensive in number, to select the prominent features and to retain their properties, correlation feature sub-pattern-based PCA (SPPCA), and cross sub-pattern correlation-based PCA (SUBXPCA) are used as a dimensionality reduction techniques. To validate the proposed work, performance evaluation parameter such as the accuracy of the time-frequency domain features from different combinations of the dataset has been compared with the latest state-of-the-art works. Simulation results show that the proposed algorithm combined with machine learning classifiers. The best accuracy of 97% for SPPCA is achieved by CatBoost and for SUBXPCA the best accuracy of 98% is achieved by random forest classifier which clearly outperformed the other related works, both in terms of accuracy and computational complexity.</pre>

scholarly journals Detection of Epileptic Seizures from EEG Signals by Combining Dimensionality Reduction Algorithms with Machine Learning Models

2021 ◽  
Author(s):  
Muhammad Zubair
Keyword(s):  
Machine Learning ◽  
Dimensionality Reduction ◽  
Frequency Domain ◽  
Work Performance ◽  
Epileptic Seizures ◽  
Dominant Frequency ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Eeg Signals ◽  
Time Frequency

<pre>In this paper, more emphasis has been given to develop a support vector machine (SVM) model using SPPCA and SUBXPCA dimensionality reduction algorithms to increase the classification accuracy. Firstly, Discrete Wavelet Transform (DWT) is applied to EEG signals for extracting the time-frequency domain features of epilepsy such as the energy of each sub-pattern, spike rhythmicity, relative spike amplitude, Dominant Frequency (DF) and Spectral Entropy (SE). The features obtained after performing DWT on an EEG signal are extensive in number, to select the prominent features and to retain their properties, correlation feature sub-pattern-based PCA (SPPCA), and cross sub-pattern correlation-based PCA (SUBXPCA) are used as a dimensionality reduction techniques. To validate the proposed work, performance evaluation parameter such as the accuracy of the time-frequency domain features from different combinations of the dataset has been compared with the latest state-of-the-art works. Simulation results show that the proposed algorithm combined with machine learning classifiers. The best accuracy of 97% for SPPCA is achieved by CatBoost and for SUBXPCA the best accuracy of 98% is achieved by random forest classifier which clearly outperformed the other related works, both in terms of accuracy and computational complexity.</pre>

scholarly journals Classification of Emotional States Inparkinson’s Disease Patients Using Time,Frequency and Time-Frequency Analysis

2021 ◽  
Author(s):  
Rejith K.N ◽  
Kamalraj Subramaniam ◽  
Ayyem Pillai Vasudevan Pillai ◽  
Roshini T V ◽  
Renjith V. Ravi ◽  
...  
Keyword(s):  
Frequency Domain ◽  
Classification Accuracy ◽  
Machine Learning Algorithms ◽  
Signal Frequency ◽  
Spectral Energy ◽  
Support Vector ◽  
Svm Classifier ◽  
Time Frequency ◽  
Energy Entropy ◽  
Teager Energy

Abstract In this work, PD patients and healthy individuals were categorized with machine-learning algorithms. EEG signals associated with six different emotions, (Happiness(E1), Sadness(E2), Fear(E3), Anger(E4), Surprise,(E5) and disgust(E6)) were used for the study. EEG data were collected from 20 PD patients and 20 normal controls using multimodal stimuli. Different features were used to categorize emotional data. Emotional recognition in Parkinson’s disease (PD) has been investigated in three domains namely, time, frequency and time frequency using Entropy, Energy-Entropy and Teager Energy-Entropy features. Three classifiers namely, K-Nearest Neighbor Algorithm, Support Vector Machine and Probabilistic Neural Network were used to observethe classification results. Emotional EEG stimuli such as anger, surprise, happiness, sadness, fear, and disgust were used to categorize PD patients and healthy controls (HC). For each EEG signal, frequency features corresponding to alpha, beta and gamma bands were obtained for nine feature extraction methods (Entropy, Energy Entropy, Teager Energy Entropy, Spectral Entropy, Spectral Energy-Entropy, Spectral Teager Energy-Entropy, STFT Entropy, STFT Energy-Entropy and STFT Teager Energy-Entropy). From the analysis, it is observed that the entropy feature in frequency domain performs evenly well (above 80 %) for all six emotions with KNN. Classification results shows that using the selected energy entropy combination feature in frequency domain provides highest accuracy for all emotions except E1 and E2 for KNN and SVM classifier, whereas other features give accuracy values of above 60% for most emotions.It is also observed that emotion E1 gives above 90 % classification accuracy for all classifiers in time domain.In frequency domain also, emotion E1 gives above 90% classification accuracy using PNN classifier.

Debonding Analysis and Identification of the Interface Between Sleeper and Track Slab for Twin-Block Slab Tracks

2021 ◽  
pp. 2140011
Author(s):  
Wei Du ◽  
Shi-Jie Deng ◽  
Juan-Juan Ren ◽  
Ze-Ming Zhao ◽  
Zhen Wei ◽  
...  
Keyword(s):  
Frequency Domain ◽  
Support Vector ◽  
Particle Swarm Algorithm ◽  
Interface Debonding ◽  
Lateral Side ◽  
Zone Model ◽  
Vertical Acceleration ◽  
Twin Block ◽  
Time Frequency ◽  
Track Slab

For China Railway Track System (CRTS) I twin-block slab tracks, the interface between the sleeper and track slab is susceptible to damage under the coupled effect of long-term train load and external environment factors. In order to analyze the damage behavior and identify the type of debonding at the interface, this paper established a three-dimensional finite element model and introduced the cohesion zone model and concrete damaged plasticity model to simulate the interface damage and the inner-layer damage of the track slab, respectively. The interface debonding induced by the temperature effect was analyzed, and the debonding types were identified based on the obtained vertical vibration responses of the sleeper surface under the train load. The results reveal that the damage mainly occurs on the bottom and lateral sides at the interface under the temperature load. The track model can be refined further to obtain higher analysis accuracy with acceptable calculation time using the sequential loading method. The 26 damage features derived from the time domain, frequency domain, and time–frequency domain are in good representativeness in reflecting the damage information hidden in the vibration signals. Among them, the peak values (maximum vertical acceleration of the sleeper) are 55.0, 56.7, 60.3, and 61.6[Formula: see text]m/s2 for no debonding, debonding on the lateral side, debonding at the bottom, and debonding on the longitudinal side of the interface under train load, respectively. Moreover, the identification accuracy of the debonding type can reach 93.75% combining the particle swarm algorithm and support vector machine. It indicates that the proposed identification method is effective and reliable to provide theoretical guidance for developing scientific maintenance and repair strategies for twin-block slab tracks.

Classification of EEG Signals for Motor Imagery Based on Mutual Information and Adaptive Neuro Fuzzy Inference System

Fuzzy Systems ◽  
2017 ◽  
pp. 347-366
Author(s):  
Shereen A. El-aal ◽  
Rabie A. Ramadan ◽  
Neveen Ghali
Keyword(s):  
Mutual Information ◽  
Motor Imagery ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Imagery Task ◽  
Eeg Signals ◽  
Inference System ◽  
Neuro Fuzzy ◽  
Anfis Classifier ◽  
Maximal Relevance

Electroencephalogram (EEG) signals based Brain Computer Interface (BCI) is employed to help disabled people to interact better with the environment. EEG signals are recorded through BCI system to translate it to control commands. There are a large body of literature targeting EEG feature extraction and classification for Motor Imagery tasks. Motor imagery task have several features can be extracted to use in classification. However, using more features consume running time and using irrelevant and redundant features affect the performance of the used classifier. This paper is dedicated to extracting the best feature vector for motor imagery task. This work suggests two feature selection methods based on Mutual Information (MI) including Minimum Redundancy Maximal Relevance (MRMR) and maximal Relevance (MaxRel). Adaptive Neuro Fuzzy Inference System (ANFIS) classifier with Subtractive clustering method is utilized for EEG signals classifications. The suggested methods are applied to BCI Competition III dataset IVa and IVb and BCI Competition II dataset III.

Classification of EEG Signals for Motor Imagery based on Mutual Information and Adaptive Neuro Fuzzy Inference System

2016 ◽  
Vol 5 (4) ◽  
pp. 64-82 ◽  
Author(s):  
Shereen A. El-aal ◽  
Rabie A. Ramadan ◽  
Neveen I. Ghali
Keyword(s):  
Mutual Information ◽  
Motor Imagery ◽  
Fuzzy Inference System ◽  
Fuzzy Inference ◽  
Subtractive Clustering ◽  
Imagery Task ◽  
Eeg Signals ◽  
Inference System ◽  
Neuro Fuzzy ◽  
Maximal Relevance

Electroencephalogram (EEG) signals based Brain Computer Interface (BCI) is employed to help disabled people to interact better with the environment. EEG signals are recorded through BCI system to translate it to control commands. There are a large body of literature targeting EEG feature extraction and classification for Motor Imagery tasks. Motor imagery task have several features can be extracted to use in classification. However, using more features consume running time and using irrelevant and redundant features affect the performance of the used classifier. This paper is dedicated to extracting the best feature vector for motor imagery task. This work suggests two feature selection methods based on Mutual Information (MI) including Minimum Redundancy Maximal Relevance (MRMR) and maximal Relevance (MaxRel). Adaptive Neuro Fuzzy Inference System (ANFIS) classifier with Subtractive clustering method is utilized for EEG signals classifications. The suggested methods are applied to BCI Competition III dataset IVa and IVb and BCI Competition II dataset III.

Sign in / Sign up

Export Citation Format

Share Document