Frequency Bands Of EEG Signals Research Articles

SFT-Net: A Network for Detecting Fatigue From EEG Signals by Combining 4D Feature Flow and Attention Mechanism.

Fatigued driving is a leading cause of traffic accidents, and accurately predicting driver fatigue can significantly reduce their occurrence. However, modern fatigue detection models based on neural networks often face challenges such as poor interpretability and insufficient input feature dimensions. This article proposes a novel Spatial-Frequency-Temporal Network (SFT-Net) method for detecting driver fatigue using electroencephalogram (EEG) data. Our approach integrates EEG signals' spatial, frequency, and temporal information to improve recognition performance. We transform the differential entropy of five frequency bands of EEG signals into a 4D feature tensor to preserve these three types of information. An attention module is then used to recalibrate the spatial and frequency information of each input 4D feature tensor time slice. The output of this module is fed into a depthwise separable convolution (DSC) module, which extracts spatial and frequency features after attention fusion. Finally, long short-term memory (LSTM) is used to extract the temporal dependence of the sequence, and the final features are output through a linear layer. We validate the effectiveness of our model on the SEED-VIG dataset, and experimental results demonstrate that SFT-Net outperforms other popular models for EEG fatigue detection. Interpretability analysis supports the claim that our model has a certain level of interpretability. Our work addresses the challenge of detecting driver fatigue from EEG data and highlights the importance of integrating spatial, frequency, and temporal information.

Predicting rock-paper-scissors choices based on single-trial EEG signals.

Decision prediction based on neurophysiological signals is of great application value in many real-life situations, especially in human-AI collaboration or counteraction. Single-trial analysis of electroencephalogram (EEG) signals is a very valuable step in the development of an online decision-prediction system. However, previous EEG-based decision-prediction methods focused mainly on averaged EEG signals of all decision-making trials to predict an individual's general decision tendency (e.g., risk seeking or aversion) over a period rather than on a specific decision response in a single trial. In the present study, we used a rock-paper-scissors game, which is a common multichoice decision-making task, to explore how to predict participants' single-trial choice with EEG signals. Forty participants, comprising 20 females and 20 males, played the game with a computer player for 330 trials. Considering that the decision-making process of this game involves multiple brain regions and neural networks, we proposed a new algorithm named common spatial pattern-attractor metagene (CSP-AM) to extract CSP features from different frequency bands of EEG signals that occurred during decision making. The results showed that a multilayer perceptron classifier achieved an accuracy significantly exceeding the chance level among 88.57% (31 of 35) of participants, verifying the classification ability of CSP features in multichoice decision-making prediction. We believe that the CSP-AM algorithm could be used in the development of proactive AI systems.

Prediction of treatment outcome for repetitive transcranial magnetic stimulation in major depressive disorder using connectivity measures and ensemble of pre-trained deep learning models

Repetitive Transcranial Magnetic Stimulation (rTMS) can be used as an effective treatment for Major Depressive Disorder (MDD) especially when a patient does not respond to multiple antidepressants. However, the prediction of the treatment outcome of rTMS is a vital task to prevent starting an inefficient treatment which may waste several important weeks for the patient and clinic. In the present study, we acquired 19-channel Electro-Encephalogram (EEG) for 34 MDD patients diagnosed with drug-resistant Depression before initiating rTMS treatment. Effective Connectivity matrix was obtained from four frequency bands of EEG signal to create connectivity images which, simultaneously, represent spatial and frequency information of EEG signals. Then, we investigated five powerful pre-trained Convolutional Neural Networks (CNN) named VGG16, Xception, InceptionResNetV2, DenseNet121, and EfficientNetB0 as Transfer Learning (TL) models to predict rTMS treatment outcome by discriminating Responders and NonResponders. VGG16 and DenseNet121 achieved the best performance with an accuracy of 89.22%. Additionally, we developed a new ensemble model based on all pre-trained CNN models using a weighted majority voting approach. The Differential Evolution Optimization (DEO) algorithm is used to find the optimal weights for the ensemble model and the mean accuracy of 92.28% was obtained. The superior performance of the optimized ensemble of these pre-trained CNN models, using effective connectivity images obtained from EEG signals, shows that the proposed method is highly capable of predicting the treatment outcome of rTMS.

Attention-Based Convolutional Recurrent Deep Neural Networks for the Prediction of Response to Repetitive Transcranial Magnetic Stimulation for Major Depressive Disorder.

Repetitive Transcranial Magnetic Stimulation (rTMS) is proposed as an effective treatment for major depressive disorder (MDD). However, because of the suboptimal treatment outcome of rTMS, the prediction of response to this technique is a crucial task. We developed a deep learning (DL) model to classify responders (R) and non-responders (NR). With this aim, we assessed the pre-treatment EEG signal of 34 MDD patients and extracted effective connectivity (EC) among all electrodes in four frequency bands of EEG signal. Two-dimensional EC maps are put together to create a rich connectivity image and a sequence of these images is fed to the DL model. Then, the DL framework was constructed based on transfer learning (TL) models which are pre-trained convolutional neural networks (CNN) named VGG16, Xception, and EfficientNetB0. Then, long short-term memory (LSTM) cells are equipped with an attention mechanism added on top of TL models to fully exploit the spatiotemporal information of EEG signal. Using leave-one subject out cross validation (LOSO CV), Xception-BLSTM-Attention acquired the highest performance with 98.86% of accuracy and 97.73% of specificity. Fusion of these models as an ensemble model based on optimized majority voting gained 99.32% accuracy and 98.34% of specificity. Therefore, the ensemble of TL-LSTM-Attention models can predict accurately the treatment outcome.

Adaptive transfer learning-based multiscale feature fused deep convolutional neural network for EEG MI multiclassification in brain–computer interface

Objective.Deep learning (DL)-based brain–computer interface (BCI) in motor imagery (MI) has emerged as a powerful method for establishing direct communication between the brain and external electronic devices. However, due to inter-subject variability, inherent complex properties, and low signal-to-noise ratio (SNR) in electroencephalogram (EEG) signals are major challenges that significantly hinder the accuracy of the MI classifier. Approach.To overcome this, the present work proposes an efficient transfer learning (TL)-based multi-scale feature fused CNN (MSFFCNN) which can capture the distinguishable features of various non-overlapping canonical frequency bands of EEG signals from different convolutional scales for multi-class MI classification. Significance.In order to account for inter-subject variability from different subjects, the current work presents 4 different model variants including subject-independent and subject-adaptive classification models considering different adaptation configurations to exploit the full learning capacity of the classifier. Each adaptation configuration has been fine-tuned in an extensively trained pre-trained model and the performance of the classifier has been studied for a vast range of learning rates and degrees of adaptation which illustrates the advantages of using an adaptive transfer learning-based model. Results.The model achieves an average classification accuracy of 94.06% (±0.70%) and the kappa value of 0.88 outperforming several baseline and current state-of-the-art EEG-based MI classification models with fewer training samples. The present research provides an effective and efficient transfer learning-based end-to-end MI classification framework for designing a high-performance robust MI-BCI system.

An efficient multi-scale CNN model with intrinsic feature integration for motor imagery EEG subject classification in brain-machine interfaces

ObjectiveElectroencephalogram (EEG) based motor imagery (MI) classification is an important aspect in brain-machine interfaces (BMIs) which bridges between neural system and computer devices decoding brain signals into recognizable machine commands. However, the MI classification task is challenging due to inherent complex properties, inter-subject variability, and low signal-to-noise ratio (SNR) of EEG signals. To overcome the above-mentioned issues, the current work proposes an efficient multi-scale convolutional neural network (MS-CNN) which can extract the distinguishable features of several non-overlapping canonical frequency bands of EEG signals from multiple scales for MI-BCI classification. ApproachIn the framework, discriminant user-specific features have been extracted and integrated to improve the accuracy and performance of the CNN classifier. Additionally, different data augmentation methods have been implemented to further improve the accuracy and robustness of the model. Main resultsThe model achieves an average classification accuracy of 93.74% and Cohen’s kappa-coefficient of 0.92 on the BCI competition IV2b dataset outperforming several baseline and current state-of-the-art EEG-based MI classification models. SignificanceThe proposed algorithm effectively addresses the shortcoming of existing CNN-based EEG-MI classification models and significantly improves the classification accuracy. The current framework can provide a stimulus for designing efficient and robust real-time human-robot interaction.

An EEG Neurofeedback Interactive Model for Emotional Classification of Electronic Music Compositions Considering Multi-Brain Synergistic Brain-Computer Interfaces.

This paper presents an in-depth study and analysis of the emotional classification of EEG neurofeedback interactive electronic music compositions using a multi-brain collaborative brain-computer interface (BCI). Based on previous research, this paper explores the design and performance of sound visualization in an interactive format from the perspective of visual performance design and the psychology of participating users with the help of knowledge from various disciplines such as psychology, acoustics, aesthetics, neurophysiology, and computer science. This paper proposes a specific mapping model for the conversion of sound to visual expression based on people’s perception and aesthetics of sound based on the phenomenon of audiovisual association, which provides a theoretical basis for the subsequent research. Based on the mapping transformation pattern between audio and visual, this paper investigates the realization path of interactive sound visualization, the visual expression form and its formal composition, and the aesthetic style, and forms a design expression method for the visualization of interactive sound, to benefit the practice of interactive sound visualization. In response to the problem of neglecting the real-time and dynamic nature of the brain in traditional brain network research, dynamic brain networks proposed for analyzing the EEG signals induced by long-time music appreciation. During prolonged music appreciation, the connectivity of the brain changes continuously. We used mutual information on different frequency bands of EEG signals to construct dynamic brain networks, observe changes in brain networks over time and use them for emotion recognition. We used the brain network for emotion classification and achieved an emotion recognition rate of 67.3% under four classifications, exceeding the highest recognition rate available.

Electroencephalogram Emotion Recognition Based on 3D Feature Fusion and Convolutional Autoencoder.

As one of the key technologies of emotion computing, emotion recognition has received great attention. Electroencephalogram (EEG) signals are spontaneous and difficult to camouflage, so they are used for emotion recognition in academic and industrial circles. In order to overcome the disadvantage that traditional machine learning based emotion recognition technology relies too much on a manual feature extraction, we propose an EEG emotion recognition algorithm based on 3D feature fusion and convolutional autoencoder (CAE). First, the differential entropy (DE) features of different frequency bands of EEG signals are fused to construct the 3D features of EEG signals, which retain the spatial information between channels. Then, the constructed 3D features are input into the CAE constructed in this paper for emotion recognition. In this paper, many experiments are carried out on the open DEAP dataset, and the recognition accuracy of valence and arousal dimensions are 89.49 and 90.76%, respectively. Therefore, the proposed method is suitable for emotion recognition tasks.

Cognitive Imagery Classification of EEG Signals using CSP-based Feature Selection Method

This paper presents a novel approach of spectral feature selection using spatial filters for the classification of four cognitive imagery tasks. The input dataset consists of electroencephalogram (EEG) signals acquired through a commercial wireless headset. The spectral features included mel frequency (MF) components extracted from the low frequency bands of EEG signal. A spatial projection filter was used for the selection of the most relevant features before classification. The popular method of multiclass common spatial pattern (CSP) and regularized CSP (RCSP) are investigated for a subject dependent (intra) and subject independent (inter) generation of spatial projection filter, respectively. Based upon this, present study used two different algorithmic approaches namely MF-CSP and MF-RCSP. The developed algorithm successfully classified four imagery actions with the reported prediction accuracy of 46.23% and 64.01% and standard deviation of 11.60% and 8.67% for MF-CSP and MF-RCSP, respectively.

Feature Construction Methods for the Electroencephalogram Signal Analysis in Hybrid “Eye-Brain-Computer” Interface

The hybrid “eye-brain-computer” interface is a new approach to the human-machine interaction. It allows the user to select an object of interest on a screen by tracking the user’s gaze direction. At the same time, the user’s intent to give a command is determined by registering and decoding brain activity. The interface operation is based on the fact that control gaze fixations can be distinguished from spontaneous fixations using electroencephalogram (EEG) signal.The article discusses the recognition of EEG patterns that correspond to the spontaneous and control gaze fixations. To improve the classification accuracy, we suggest using the relatively new feature construction methods for time series analysis. These methods include a selection of optimal frequency bands of the multivariate EEG signal and a modified method of shapelets. The first method constructs the optimal feature space using prior information on a difference in frequency components of the multivariate signal for different classes. The second method uses a genetic algorithm to provide selecting such fragments of the multivariate time-series, which reflect as much as possible the properties of one or more than one class of such time series. Thus, calculating distances between them and a set of k top-best shapelets allows us to provide feature description of the time series.The article consists of five sections. The first one provides a mathematical formulation of the multivariate time-series classification problem. The second section gives a formal description of the proposed methods for feature construction. The third section describes test data, which include the EEG records from the six users of the hybrid “eye-brain-computer” interface. In the fourth section, we evaluate an efficiency of the methods proposed in comparison with other known feature extraction techniques, which include: 1) calculation of the average EEG amplitude values in the overlapping windows; 2) estimation of the power spectral density in the specified frequency bands; 3) selection of the most informative features using a genetic algorithm. In the fifth section, we conduct the statistical analysis of the results obtained. It is shown that the feature construction method, based on the selection of optimal frequency bands of the EEG signal, in efficiency significantly outperforms other techniques considered and opens up the possibility to reduce the number of false positives of the hybrid interface.

Rhythm‐based features for classification of focal and non‐focal EEG signals

Electroencephalogram (EEG) contains five rhythms, which provide details about various activities of brain. These rhythms are separated using Hilbert-Huang transform for classification of focal and non-focal EEG signals. For this, the EEG signal is disintegrated into narrow bands intrinsic mode functions (IMFs) using empirical mode decomposition, and analytic representation of IMFs is computed by Hilbert transformation that helps to extract instantaneous frequencies of respective IMFs. Frequency bands of EEG signals known as rhythms are separated from analytic IMFs using instantaneous frequencies. Two efficient parameters Pearson product-moment correlation coefficient and Spearman rank correlation coefficient extracted from the rhythms are used with different kernel functions of least-squares support vector machine for the classification of focal and non-focal EEG signals. Thus, obtained results show improved performance of proposed method as compared to other existing methods.

Dynamic Changes in Spectral and Spatial Signatures of High Frequency Oscillations in Rat Hippocampi during Epileptogenesis in Acute and Chronic Stages.

To analyze spectral and spatial signatures of high frequency oscillations (HFOs), which include ripples and fast ripples (FRs, >200 Hz) by quantitatively assessing average and peak spectral power in a rat model of different stages of epileptogenesis. The lithium-pilocarpine model of temporal lobe epilepsy was used. The acute phase of epilepsy was assessed by recording intracranial electroencephalography (EEG) activity for 1 day after status epilepticus (SE). The chronic phase of epilepsy, including spontaneous recurrent seizures (SRSs), was assessed by recording EEG activity for 28 days after SE. Average and peak spectral power of five frequency bands of EEG signals in CA1, CA3, and DG regions of the hippocampus were analyzed with wavelet and digital filter. FRs occurred in the hippocampus in the animal model. Significant dynamic changes in the spectral power of FRS were identified in CA1 and CA3. The average spectral power of ripples increased at 20 min before SE (p < 0.05), peaked at 10 min before diazepam injection. It decreased at 10 min after diazepam (p < 0.05) and returned to baseline after 1 h. The average spectral power of FRs increased at 30 min before SE (p < 0.05) and peaked at 10 min before diazepam. It decreased at 10 min after diazepam (p < 0.05) and returned to baseline at 2 h after injection. The dynamic changes were similar between average and peak spectral power of FRs. Average and peak spectral power of both ripples and FRs in the chronic phase showed a gradual downward trend compared with normal rats 14 days after SE. The spectral power of HFOs may be utilized to distinguish between normal and pathologic HFOs. Ictal average and peak spectral power of FRs were two parameters for predicting acute epileptic seizures, which could be used as a new quantitative biomarker and early warning marker of seizure. Changes in interictal HFOs power in the hippocampus at the chronic stage may be not related to seizure occurrence.

Wavelet-based emotion recognition system using EEG signal

In this research, emotional states in arousal/valence dimensions have been classified using minimum number of channels and frequency bands of EEG signal. Using the discrete wavelet transforms, EEG signals have been decomposed to corresponding frequency bands and then several features have been extracted. The support vector machine and K-nearest neighbor classifiers have been used to detect the emotional states from the extracted features. For the recorded 10-channel EEG signal, results illustrate the classification accuracy of 86.75 % for arousal level and 84.05 % for valence level. Moreover, using the high-frequency bands, specifically gamma band, yields higher accuracy compared to using low-frequency bands of EEG signal. All of these support to the development of a real-time emotion classification system.

CLASSIFICATION OF EEG SIGNALS IN NORMAL AND DEPRESSION CONDITIONS BY ANN USING RWE AND SIGNAL ENTROPY

EEG is useful for the analysis of the functional activity of the brain and a detailed assessment of this non-stationary waveform can provide crucial parameters indicative of the mental state of patients. The complex nature of EEG signals calls for automated analysis using various signal processing methods. This paper attempts to classify the EEG signals of normal and depression patients using well-established signal processing techniques involving relative wavelet energy (RWE) and artificial feedForward neural network. High frequency noise present in the recorded signal is removed using total variation filtering (TVF). Classification of the frequency bands of EEG signals into appropriate detail levels and approximation level is carried out using an eight-level multiresolution decomposition method of discrete wavelet transform (DWT). Parseval's theorem is used for calculating the energy at different resolution levels. RWE analysis gives information about the signal energy distribution at different decomposition levels. Both RWE and feedforward Network are used to classify the signals from normal controls and depression patients. The performance of the artificial neural network was evaluated using the classification accuracy and its value of 98.11% indicates a great potential for classifying normal and depression signals.