Raw EEG Research Articles

Lightweight convolution transformer for cross-patient seizure detection in multi-channel EEG signals

BackgroundEpilepsy is a neurological illness affecting the brain that makes people more likely to experience frequent, spontaneous seizures. There has to be an accurate automated method for measuring seizures frequency and severity to assess the efficacy of pharmacological therapy for epilepsy. The drug quantities are often derived from patient reports which may cause significant issues owing to inadequate or inaccurate descriptions of seizures and their frequencies. Methods and materialsThis study proposes a novel deep learning architecture-based Lightweight Convolution Transformer (LCT). The Transformer model is able to learn spatial and temporal correlated information simultaneously from the multi-channel electroencephalogram (EEG) signal to detect seizures at smaller segment lengths. In the proposed work, the lack of translation equivariance and localization of ViT is reduced using convolution tokenization, and rich information from the Transformer encoder is extracted by sequence pooling instead of the learnable class token. ResultsExtensive experimental results demonstrate that the proposed model on cross-patient learning can effectively detect seizures from the raw EEG signals. The accuracy and F1-score of seizure detection in the cross-patient case on the CHB-MIT dataset are 96.31% and 96.32%, respectively, at 0.5 sec segment length. In addition, the performance metrics show that the inclusion of inductive biases and attention-based pooling in the model enhances the performance and reduces the number of Transformer encoder layers, which significantly reduces the computational complexity. In this research, we provide a novel approach to enhance efficiency and simplify the architecture for multi-channel automated seizure detection.

Is the Patient State Index a reliable parameter as guide to anaesthesiology in cranial neurosurgery? A first intraoperative study and a literature review

BackgroundPatient State Index (PSI) and Suppression Ratio (SR) are two indices calculated by quantitative analysis of EEG used to estimate the depth of anaesthesia but their validation in neurosurgery must be done. Our aim was to investigate the congruity PSI and SR with raw EEG monitoring in neurosurgery. MethodsWe included 34 patients undergoing elective cranial neurosurgery. Each patient was monitored by a SedLine device (PSI and SR) and by raw EEG. To appraise the agreement between PSI, SR and EEG Suppr%, Bland-Altman analysis was used. We also correlated the PSI and SR recorded at different times during surgery to the degree of suppression of the raw EEG data by Spearman's rank correlation coefficient. For a comparison with previous data we made an international literature review according to PRISMA protocol. ResultsAt all recording times, we found that there is a strong agreement between PSI and raw EEG. We also found a significant correlation for both PSI and SR with the EEG suppression percentage (p < 0.05), but with a broad dispersion of the individual values within the confidence interval. ConclusionThe Masimo SedLine processed EEG monitoring system can be used as a guide in the anaesthetic management of patients during elective cranial neurosurgery, but the anaesthesiologist must be aware that previous correlations between PSI and SR with the suppression percentage may not always be valid in all individual patients. The use of an extended visual raw EEG evaluated by an expert electroencephalographer might help to provide better guidance.

Multi-task learning for arousal and sleep stage detection using fully convolutional networks

Objective. Sleep is a critical physiological process that plays a vital role in maintaining physical and mental health. Accurate detection of arousals and sleep stages is essential for the diagnosis of sleep disorders, as frequent and excessive occurrences of arousals disrupt sleep stage patterns and lead to poor sleep quality, negatively impacting physical and mental health. Polysomnography is a traditional method for arousal and sleep stage detection that is time-consuming and prone to high variability among experts. Approach. In this paper, we propose a novel multi-task learning approach for arousal and sleep stage detection using fully convolutional neural networks. Our model, FullSleepNet, accepts a full-night single-channel EEG signal as input and produces segmentation masks for arousal and sleep stage labels. FullSleepNet comprises four modules: a convolutional module to extract local features, a recurrent module to capture long-range dependencies, an attention mechanism to focus on relevant parts of the input, and a segmentation module to output final predictions. Main results. By unifying the two interrelated tasks as segmentation problems and employing a multi-task learning approach, FullSleepNet achieves state-of-the-art performance for arousal detection with an area under the precision-recall curve of 0.70 on Sleep Heart Health Study and Multi-Ethnic Study of Atherosclerosis datasets. For sleep stage classification, FullSleepNet obtains comparable performance on both datasets, achieving an accuracy of 0.88 and an F1-score of 0.80 on the former and an accuracy of 0.83 and an F1-score of 0.76 on the latter. Significance. Our results demonstrate that FullSleepNet offers improved practicality, efficiency, and accuracy for the detection of arousal and classification of sleep stages using raw EEG signals as input.

An ensemble deep-learning approach for single-trial EEG classification of vibration intensity

Objective. Single-trial electroencephalography (EEG) classification is a promising approach to evaluate the cognitive experience associated with haptic feedback. Convolutional neural networks (CNNs), which are among the most widely used deep learning techniques, have demonstrated their effectiveness in extracting EEG features for the classification of different cognitive functions, including the perception of vibration intensity that is often experienced during human-computer interaction. This paper proposes a novel CNN ensemble model to classify the vibration-intensity from a single trial EEG data that outperforms the state-of-the-art EEG models. Approach. The proposed ensemble model, named SE NexFusion, builds upon the observed complementary learning behaviors of the EEGNex and TCNet Fusion models, exhibited in learning personal as well generic neural features associated with vibration intensity. The proposed ensemble employs multi-branch feature encoders corroborated with squeeze-and-excitation units that enables rich-feature encoding while at the same time recalibrating the weightage of the obtained feature maps based on their discriminative power. The model takes in a single trial of raw EEG as an input and does not require complex EEG signal-preprocessing. Main results. The proposed model outperforms several state-of-the-art bench-marked EEG models by achieving an average accuracy of 60.7% and 61.6% under leave-one-subject-out and within-subject cross-validation (three-classes), respectively. We further validate the robustness of the model through Shapley values explainability method, where the most influential spatio-temporal features of the model are counter-checked with the neural correlates that encode vibration intensity. Significance. Results show that SE NexFusion outperforms other benchmarked EEG models in classifying the vibration intensity. Additionally, explainability analysis confirms the robustness of the model in attending to features associated with the neural correlates of vibration intensity.

Automatic sleep staging by a hybrid model based on deep 1D-ResNet-SE and LSTM with single-channel raw EEG signals.

Sleep staging is crucial for assessing sleep quality and diagnosing sleep disorders. Recent advances in deep learning methods with electroencephalogram (EEG) signals have shown remarkable success in automatic sleep staging. However, the use of deeper neural networks may lead to the issues of gradient disappearance and explosion, while the non-stationary nature and low signal-to-noise ratio of EEG signals can negatively impact feature representation. To overcome these challenges, we proposed a novel lightweight sequence-to-sequence deep learning model, 1D-ResNet-SE-LSTM, to classify sleep stages into five classes using single-channel raw EEG signals. Our proposed model consists of two main components: a one-dimensional residual convolutional neural network with a squeeze-and-excitation module to extract and reweight features from EEG signals, and a long short-term memory network to capture the transition rules among sleep stages. In addition, we applied the weighted cross-entropy loss function to alleviate the class imbalance problem. We evaluated the performance of our model on two publicly available datasets; Sleep-EDF Expanded consists of 153 overnight PSG recordings collected from 78 healthy subjects and ISRUC-Sleep includes 100 PSG recordings collected from 100 subjects diagnosed with various sleep disorders, and obtained an overall accuracy rate of 86.39% and 81.97%, respectively, along with corresponding macro average F1-scores of 81.95% and 79.94%. Our model outperforms existing sleep staging models in terms of overall performance metrics and per-class F1-scores for several sleep stages, particularly for the N1 stage, where it achieves F1-scores of 59.00% and 55.53%. The kappa coefficient is 0.812 and 0.766 for the Sleep-EDF Expanded and ISRUC-Sleep datasets, respectively, indicating strong agreement with certified sleep experts. We also investigated the effect of different weight coefficient combinations and sequence lengths of EEG epochs used as input to the model on its performance. Furthermore, the ablation study was conducted to evaluate the contribution of each component to the model's performance. The results demonstrate the effectiveness and robustness of the proposed model in classifying sleep stages, and highlights its potential to reduce human clinicians' workload, making sleep assessment and diagnosis more effective. However, the proposed model is subject to several limitations. Firstly, the model is a sequence-to-sequence network, which requires input sequences of EEG epochs. Secondly, the weight coefficients in the loss function could be further optimized to balance the classification performance of each sleep stage. Finally, apart from the channel attention mechanism, incorporating more advanced attention mechanisms could enhance the model's effectiveness.

Efficacy of transformer networks for classification of EEG data

With the unprecedented success of transformer networks in natural language processing (NLP), recently, they have been successfully adapted to areas like computer vision, generative adversarial networks (GAN), and reinforcement learning. Classifying electroencephalogram (EEG) data has been challenging and researchers have been overly dependent on pre-processing and hand-crafted feature extraction. Despite having achieved automated feature extraction in several other domains, deep learning has not yet been accomplished for EEG. In this paper, the efficacy of the transformer network for the classification of raw EEG data (cleaned and pre-processed) is explored. The performance of transformer networks was evaluated on a local (age and gender data) and a public dataset (STEW). First, a classifier using a transformer network is built to classify the age and gender of a person with raw resting-state EEG data. Second, the classifier is tuned for mental workload classification with open access raw multi-tasking mental workload EEG data (STEW). The network achieves an accuracy comparable to state-of-the-art accuracy on both the local (Age and Gender dataset; 94.53% (gender) and 87.79% (age)) and the public (STEW dataset; 95.28% (two workload levels) and 88.72% (three workload levels)) dataset. The accuracy values have been achieved using raw EEG data without feature extraction. Results indicate that the transformer-based deep learning models can successfully abate the need for heavy feature-extraction of EEG data for successful classification.

A multi-dimensional hybrid CNN-BiLSTM framework for epileptic seizure detection using electroencephalogram signal scrutiny

The proposed hybrid CNN-BiLSTM architecture aims to address the challenge of detecting epileptic seizures systematically from EEG signal analysis. The system consists of several stages, including preprocessing, feature extraction using multi-dimensional CNN, temporal feature processing using BiLSTM, and classification using fully connected layers. The first stage involves preprocessing and normalization of the raw EEG signal to prepare it for further analysis. This step helps in removing noise and standardizing the input for subsequent processing. Next, a multi-dimensional CNN is employed to effectively extract features from the preprocessed EEG sequence data. CNNs are known for their ability to capture spatial features, and in this case, they are utilized to extract relevant features from the EEG data. After the feature extraction stage, the BiLSTM component of the architecture is utilized to process the extracted features and capture temporal dependencies. BiLSTMs are well-suited for sequence modeling tasks and can effectively capture long-range dependencies in the data. By incorporating BiLSTM, the architecture aims to capture important temporal patterns related to epileptic seizures. Finally, the temporal feature values are fed into fully connected layers for classification. The system is designed to detect epileptic seizures and classify them into specific types using a 10-class classification approach. The proposed system reports high detection accuracy with an overall accuracy of 99.53% and an accuracy of 82.95% on the binary classification task. These results suggest that the system performs well in accurately identifying epileptic seizures from EEG signals. Furthermore, the proposed system demonstrates superior performance compared to other existing techniques such as K-nearest neighbor (KNN), support vector machine (SVM), and decision tree (DT) in terms of accuracy. It is important to note that the evaluation and comparison of the proposed system were performed on a publicly available epileptic seizures image database. Overall, the proposed hybrid CNN-BiLSTM architecture shows potential in enhancing the detection and classification of epileptic seizures from EEG signals, potentially improving the efficiency and accuracy of diagnosis and treatment in the early stages of these disorders.

Electroencephalogram-Based Subject Matching Learning (ESML): A Deep Learning Framework on Electroencephalogram-Based Biometrics and Task Identification.

An EEG signal (Electroencephalogram) is a bioelectric phenomenon reflecting human brain activities. In this paper, we propose a novel deep learning framework ESML (EEG-based Subject Matching Learning) using raw EEG signals to learn latent representations for EEG-based user identification and tack classification. ESML consists of two parts: one is the ESML1 model via an LSTM-based method for EEG-user linking, and one is the ESML2 model via a CNN-based method for EEG-task linking. The new model ESML is simple, but effective and efficient. It does not require any restrictions for EEG data collection on motions and thinking for users, and it does not need any EEG preprocessing operations, such as EEG denoising and feature extraction. The experiments were conducted on three public datasets and the results show that ESML performs the best and achieves significant performance improvement when compared to baseline methods (i.e., SVM, LDA, NN, DTS, Bayesian, AdaBoost and MLP). The ESML1 model provided the best precision at 96% with 109 users and the ESML2 model achieved 99% precision at 3-Class task classification. These experimental results provide direct evidence that EEG signals can be used for user identification and task classification.

REM Sleep Stage Identification with Raw Single-Channel EEG.

This paper focused on creating an interpretable model for automatic rapid eye movement (REM) and non-REM sleep stage scoring for a single-channel electroencephalogram (EEG). Many methods attempt to extract meaningful information to provide to a learning algorithm. This method attempts to let the model extract the meaningful interpretable information by providing a smaller number of time-invariant signal filters for five frequency ranges using five CNN algorithms. A bi-directional GRU algorithm was applied to the output to incorporate time transition information. Training and tests were run on the well-known sleep-EDF-expanded database. The best results produced 97% accuracy, 93% precision, and 89% recall.

Transformers for autonomous recognition of psychiatric dysfunction via raw and imbalanced EEG signals

Early identification of mental disorders, based on subjective interviews, is extremely challenging in the clinical setting. There is a growing interest in developing automated screening tools for potential mental health problems based on biological markers. Here, we demonstrate the feasibility of an AI-powered diagnosis of different mental disorders using EEG data. Specifically, this work aims to classify different mental disorders in the following ecological context accurately: (1) using raw EEG data, (2) collected during rest, (3) during both eye open, and eye closed conditions, (4) at short 2-min duration, (5) on participants with different psychiatric conditions, (6) with some overlapping symptoms, and (7) with strongly imbalanced classes. To tackle this challenge, we designed and optimized a transformer-based architecture, where class imbalance is addressed through focal loss and class weight balancing. Using the recently released TDBRAIN dataset (n= 1274 participants), our method classifies each participant as either a neurotypical or suffering from major depressive disorder (MDD), attention deficit hyperactivity disorder (ADHD), subjective memory complaints (SMC), or obsessive–compulsive disorder (OCD). We evaluate the performance of the proposed architecture on both the window-level and the patient-level. The classification of the 2-min raw EEG data into five classes achieved a window-level accuracy of 63.2% and 65.8% for open and closed eye conditions, respectively. When the classification is limited to three main classes (MDD, ADHD, SMC), window level accuracy improved to 75.1% and 69.9% for eye open and eye closed conditions, respectively. Our work paves the way for developing novel AI-based methods for accurately diagnosing mental disorders using raw resting-state EEG data.

Common spatial pattern for classification of loving kindness meditation EEG for single and multiple sessions

While a very few studies have been conducted on classifying loving kindness meditation (LKM) and non-meditation electroencephalography (EEG) data for a single session, there are no such studies conducted for multiple session EEG data. Thus, this study aims at classifying existing raw EEG meditation data on single and multiple sessions to come up with meaningful inferences which will be highly beneficial when developing algorithms that can support meditation practices. In this analysis, data have been collected on Pre-Resting (before-meditation), Post-Resting (after-meditation), LKM-Self and LKM-Others for 32 participants and hence allowing us to conduct six pairwise comparisons for the four mind tasks. Common Spatial Patterns (CSP) is a feature extraction method widely used in motor imaginary brain computer interface (BCI), but not in meditation EEG data. Therefore, using CSP in extracting features from meditation EEG data and classifying meditation/non-meditation instances, particularly for multiple sessions will create a new path in future meditation EEG research. The classification was done using Linear Discriminant Analysis (LDA) where both meditation techniques (LKM-Self and LKM-Others) were compared with Pre-Resting and Post-Resting instances. The results show that for a single session of 32 participants, around 99.5% accuracy was obtained for classifying meditation/Pre-Resting instances. For the 15 participants when using five sessions of EEG data, around 83.6% accuracy was obtained for classifying meditation/Pre-Resting instances. The results demonstrate the ability to classify meditation/Pre-Resting data. Most importantly, this classification is possible for multiple session data as well. In addition to this, when comparing the classification accuracies of the six mind task pairs; LKM-Self, LKM-Others and Post-Resting produced relatively lower accuracies among them than the accuracies obtained for classifying Pre-Resting with the other three. This indicates that Pre-Resting has some features giving a better classification indicating that it is different from the other three mind tasks.

A novel framework for classification of two-class motor imagery EEG signals using logistic regression classification algorithm.

Robotics and artificial intelligence have played a significant role in developing assistive technologies for people with motor disabilities. Brain-Computer Interface (BCI) is a communication system that allows humans to communicate with their environment by detecting and quantifying control signals produced from different modalities and translating them into voluntary commands for actuating an external device. For that purpose, classification the brain signals with a very high accuracy and minimization of the errors is of profound importance to the researchers. So in this study, a novel framework has been proposed to classify the binary-class electroencephalogram (EEG) data. The proposed framework is tested on BCI Competition IV dataset 1 and BCI Competition III dataset 4a. Artifact removal from EEG data is done through preprocessing, followed by feature extraction for recognizing discriminative information in the recorded brain signals. Signal preprocessing involves the application of independent component analysis (ICA) on raw EEG data, accompanied by the employment of common spatial pattern (CSP) and log-variance for extracting useful features. Six different classification algorithms, namely support vector machine, linear discriminant analysis, k-nearest neighbor, naïve Bayes, decision trees, and logistic regression, have been compared to classify the EEG data accurately. The proposed framework achieved the best classification accuracies with logistic regression classifier for both datasets. Average classification accuracy of 90.42% has been attained on BCI Competition IV dataset 1 for seven different subjects, while for BCI Competition III dataset 4a, an average accuracy of 95.42% has been attained on five subjects. This indicates that the model can be used in real time BCI systems and provide extra-ordinary results for 2-class Motor Imagery (MI) signals classification applications and with some modifications this framework can also be made compatible for multi-class classification in the future.

Feature learning framework based on EEG graph self-attention networks for motor imagery BCI systems

Learning distinguishable features from raw EEG signals is crucial for accurate classification of motor imagery (MI) tasks. To incorporate spatial relationships between EEG sources, we developed a feature set based on an EEG graph. In this graph, EEG channels represent the nodes, with power spectral density (PSD) features defining their properties, and the edges preserving the spatial information. We designed an EEG based graph self-attention network (EGSAN) to learn low-dimensional embedding vector for EEG graph, which can be used as distinguishable features for motor imagery task classification. We evaluated our EGSAN model on two publicly available MI EEG datasets, each containing different types of motor imagery tasks. Our experiments demonstrate that our proposed model effectively extracts distinguishable features from EEG graphs, achieving significantly higher classification accuracies than existing state-of-the-art methods.

Classification of Motor Imagery Using Trial Extension in Spatial Domain with Rhythmic Components of EEG

Electrical activities of the human brain can be recorded with electroencephalography (EEG). To characterize motor imagery (MI) tasks for brain–computer interface (BCI) implementation is an easy and cost-effective tool. The MI task is represented by a short-time trial of multichannel EEG. In this paper, the signal of each channel of raw EEG is decomposed into a finite set of narrowband signals using a Fourier-transformation-based bandpass filter. Rhythmic components of EEG are represented by each of the narrowband signals that characterize the brain activities related to MI tasks. The subband signals are arranged to extend the dimension of the EEG trial in the spatial domain. The spatial features are extracted from the set of extended trials using a common spatial pattern (CSP). An optimum number of features are employed to classify the motor imagery tasks using an artificial neural network. An integrated approach with full-band and narrowband signals is implemented to derive discriminative features for MI classification. In addition, the subject-dependent parameter optimization scheme enhances the performance of the proposed method. The performance evaluation of the proposed method is obtained using two publicly available benchmark datasets (Dataset I and Dataset II). The experimental results in terms of classification accuracy (93.88% with Dataset I and 91.55% with Dataset II) show that it performs better than the recently developed algorithms. The enhanced MI classification accuracy is very much applicable in BCI implementation.

PolyCosGraph: A Privacy-Preserving Cancelable EEG Biometric System

Recent findings confirm that biometric templates derived from electroencephalography (EEG) signals contain sensitive information about registered users, such as age, gender, cognitive ability, mental status and health information. Existing privacy-preserving methods such as hash function and fuzzy commitment are not cancelable, where raw biometric features are vulnerable to hill-climbing attacks. To address this issue, we propose the PolyCosGraph, a system based on <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Poly</b> nomial transformation embedding <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cos</b> ine functions with <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Graph</b> features of EEG signals, which is a privacy-preserving and cancelable template design that protects EEG features and system security against multiple attacks. In addition, a template corrupting process is designed to further enhance the security of the system, and a corresponding matching algorithm is developed. Even when the transformed template is compromised, attackers cannot retrieve raw EEG features and the compromised template can be revoked. The proposed system achieves the authentication performance of 1.49% EER with a resting state protocol, 0.68% EER with a motor imagery task, and 0.46% EER under a watching movie condition, which is equivalent to that in the non-encrypted domain. Security analysis demonstrates that our system is resistant to attacks via record multiplicity, preimage attacks, hill-climbing attacks, second attacks and brute force attacks.

Subject Dependent and Subject Independent Analysis for Emotion Recognition Using Electroencephalogram (EEG) Signal

Brain signals for the human-computer interface is a research interest in recent years. The brain is the most vital part of our body. It handles and manages all types of activities of the body. Brain signals appear when neurons inside the brain send electrical impulses to communicate and elicit electrical potentials. This electrical activity can be measured by Electroencephalogram (EEG) through electrodes. EEG signals can help to recognize human emotions effectively. It is a non-invasive method to collect brain signals. In this paper, we have studied the subject-dependent and subject-independent analysis for four emotions (happy, sad, fear, and neutral) using the SEED-IV dataset of EEG signals for emotion. The raw EEG signals of the SEED-IV dataset have been preprocessed to remove unwanted signals and noise. 32 statistical features have been extracted from the preprocessed EEG signals and used as input for classifiers. Here, we achieved an average of 95.73% accuracy for 15 subjects for subject-dependent analysis for emotional classification using a cubic support vector machine (SVM). Based on cubic SVM and fine Gaussian SVM, we achieved an average classification accuracy of 78.46% and 83.7% for subject-independent analysis.

An eye-tracking-with-EEG coregistration corpus of narrative sentences

We present the Radboud Coregistration Corpus of Narrative Sentences (RaCCooNS), the first freely available corpus of eye-tracking-with-EEG data collected while participants read narrative sentences in Dutch. The corpus is intended for studying human sentence comprehension and for evaluating the cognitive validity of computational language models. RaCCooNS contains data from 37 participants (3 of which eye tracking only) reading 200 Dutch sentences each. Less predictable words resulted in significantly longer reading times and larger N400 sizes, replicating well-known surprisal effects in eye tracking and EEG simultaneously. We release the raw eye-tracking data, the preprocessed eye-tracking data at the fixation, word, and trial levels, the raw EEG after merger with eye-tracking data, and the preprocessed EEG data both before and after ICA-based ocular artifact correction.

Temporal transformer-spatial graph convolutional network: an intelligent classification model for anti N-methyl-D-aspartate receptor encephalitis based on electroencephalogram signal.

Encephalitis is a disease typically caused by viral infections or autoimmunity. The most common type of autoimmune encephalitis is anti-N-methyl-D-aspartate receptor (NMDAR) antibody-mediated, known as anti-NMDA receptor encephalitis, which is a rare disease. Specific EEG patterns, including "extreme delta brush" (EDB), have been reported in patients with anti-NMDA receptor encephalitis. The aim of this study was to develop an intelligent diagnostic model for encephalitis based on EEG signals. A total of 131 Participants were selected based on reasonable inclusion criteria and divided into three groups: health control (35 participants), viral encephalitis (58 participants), and anti NMDAR receptor encephalitis (55 participants). Due to the low prevalence of anti-NMDAR receptor encephalitis, it took several years to collect participants' EEG signals while they were in an awake state. EEG signals were collected and analyzed following the international 10-20 system layout. We proposed a model called Temporal Transformer-Spatial Graph Convolutional Network (TT-SGCN), which consists of a Preprocess Module, a Temporal Transformer Module (TTM), and a Spatial Graph Convolutional Module (SGCM). The raw EEG signal was preprocessed according to traditional procedures, including filtering, averaging, and Independent Component Analysis (ICA) method. The EEG signal was then segmented and transformed using short-time Fourier transform (STFT) to produce concatenated power density (CPD) maps, which served as inputs for the proposed model. TTM extracted the time-frequency features of each channel, and SGCM fused these features using graph convolutional methods based on the location of electrodes. The model was evaluated in two experiments: classification of the three groups and pairwise classification among the three groups. The model was trained using two stages and achieved the performance, with an accuracy of 82.23%, recall of 80.75%, precision of 82.51%, and F1 score of 81.23% in the classification of the three groups. The proposed model has the potential to become an intelligent auxiliary diagnostic tool for encephalitis.

Quantification of event related brain patterns for the motor imagery tasks using inter-trial variance technique

Quantification of event-related (de) synchronization (ERD/ERS) patterns is a challenging task in the field of motor imagery (MI)-based brain–computer interface (BCI). Accurately determining the optimal time and frequency band for localizing the ERD/ERS brain patterns is crucial in designing a robust MI-based BCI. To address above issue, this study proposes an inter-trial variance (IV) technique that focuses on localizing the ERD/ERS brain patterns and classifying the left and right hand imaginations of the subjects. The effectiveness of the proposed technique is validated using the BCI Competition-IV, Dataset-2b, which includes raw EEG signals from nine healthy subjects performing both left and right hand MI movements. In this technique, the sensorimotor frequency band (8–30 Hz) and its ERD/ERS brain patterns related to the left and right hand MI movements are derived from the signals. The obtained brain patterns are fed into various ML models, and the efficiency of the models is assessed using the classification accuracy (%CA) and Cohen’s kappa coefficient (K). The outcomes show that the proposed technique enhances the performance of the BCI system, classifying both left and right hand imaginations of the subjects with higher %CA (86.11%) and K (0.72), and outperforming the state-of-the-art techniques. This concludes that the ERD and ERS brain patterns extracted from the proposed technique are significant features in designing the MI-based BCI system.

Deep extreme learning machine with knowledge augmentation for EEG seizure signal recognition.

Intelligent recognition of electroencephalogram (EEG) signals can remarkably improve the accuracy of epileptic seizure prediction, which is essential for epileptic diagnosis. Extreme learning machine (ELM) has been applied to EEG signals recognition, however, the artifacts and noises in EEG signals have a serious effect on recognition efficiency. Deep learning is capable of noise resistance, contributing to removing the noise in raw EEG signals. But traditional deep networks suffer from time-consuming training and slow convergence. Therefore, a novel deep learning based ELM (denoted as DELM) motivated by stacking generalization principle is proposed in this paper. Deep extreme learning machine (DELM) is a hierarchical network composed of several independent ELM modules. Augmented EEG knowledge is taken as complementary component, which will then be mapped into next module. This learning process is so simple and fast, meanwhile, it can excavate the implicit knowledge in raw data to a greater extent. Additionally, the proposed method is operated in a single-direction manner, so there is no need to perform parameters fine-tuning, which saves the expense of time. Extensive experiments are conducted on the public Bonn EEG dataset. The experimental results demonstrate that compared with the commonly-used seizure prediction methods, the proposed DELM wins the best average accuracies in 13 out of the 22 data and the best average F-measure scores in 10 out of the 22 data. And the running time of DELM is more than two times quickly than deep learning methods. Therefore, DELM is superior to traditional and some state-of-the-art machine learning methods. The proposed architecture demonstrates its feasibility and superiority in epileptic EEG signal recognition. The proposed less computationally intensive deep classifier enables faster seizure onset detection, which is showing great potential on the application of real-time EEG signal classification.