Human EEG Signals Research Articles

Thermally Conductive and UV-EMI Shielding Electronic Textiles for Unrestricted and Multifaceted Health Monitoring

HighlightsAg nanoparticles are evenly plated on BN-embedded fibers as a thermally conductive sheath, bridging the insulating interface between fibers to construct a 3D heat transfer network.The LM-printed e-textile shows outstanding capability for the monitoring of human ECG, sEMG, and EEG signals even under intense EM interference, when commercial electrodes cannot work properly.The e-textile simultaneously manifests excellent EMI shielding (SET > 65, X-band) and UV protection (UPF = 143.1) performance, thus protecting the device and skin from harmful radiation.

MEET: A Multi-Band EEG Transformer for Brain States Decoding.

Electroencephalography (EEG) is among the most widely used and inexpensive neuroimaging techniques. Compared to the CNN or RNN based models, Transformer can better capture the temporal information in EEG signals and focus more on global features of the brain's functional activities. Importantly, according to the multiscale nature of EEG signals, it is crucial to consider the multi-band concept into the design of EEG Transformer architecture. We propose a novel Multi-band EEG Transformer (MEET) to represent and analyze the multiscale temporal time series of human brain EEG signals. MEET mainly includes three parts: 1) transform the EEG signals into multi-band images, and preserve the 3D spatial information between electrodes; 2) design a Band Attention Block to compute the attention maps of the stacked multi-band images and infer the fused feature maps; 3) apply the Temporal Self-Attention and Spatial Self-Attention modules to extract the spatiotemporal features for the characterization and differentiation of multi-frame dynamic brain states. The experimental results show that: 1) MEET outperforms state-of-the-art methods on multiple open EEG datasets (SEED, SEED-IV, WM) for brain states classification; 2) MEET demonstrates that 5-bands fusion is the best integration strategy; and 3) MEET identifies interpretable brain attention regions. MEET is an interpretable and universal model based on the multiband-multiscale characteristics of EEG. The innovative combination of band attention and temporal/spatial self-attention mechanisms in MEET achieves promising data-driven learning of the temporal dependencies and spatial relationships of EEG signals across the entire brain in a holistic and comprehensive fashion.

Exploring the Cognitive Neural Basis of Factuality in Abstractive Text Summarization Models: Interpretable Insights from EEG Signals

(1) Background: Information overload challenges decision-making in the Industry 4.0 era. While Natural Language Processing (NLP), especially Automatic Text Summarization (ATS), offers solutions, issues with factual accuracy persist. This research bridges cognitive neuroscience and NLP, aiming to improve model interpretability. (2) Methods: This research examined four fact extraction techniques: dependency relation, named entity recognition, part-of-speech tagging, and TF-IDF, in order to explore their correlation with human EEG signals. Representational Similarity Analysis (RSA) was applied to gauge the relationship between language models and brain activity. (3) Results: Named entity recognition showed the highest sensitivity to EEG signals, marking the most significant differentiation between factual and non-factual words with a score of −0.99. The dependency relation followed with −0.90, while part-of-speech tagging and TF-IDF resulted in 0.07 and −0.52, respectively. Deep language models such as GloVe, BERT, and GPT-2 exhibited noticeable influences on RSA scores, highlighting the nuanced interplay between brain activity and these models. (4) Conclusions: Our findings emphasize the crucial role of named entity recognition and dependency relations in fact extraction and demonstrate the independent effects of different models and TOIs on RSA scores. These insights aim to refine algorithms to reflect human text processing better, thereby enhancing ATS models’ factual integrity.

Human–Computer Interaction Multi-Task Modeling Based on Implicit Intent EEG Decoding

In the short term, a fully autonomous level of machine intelligence cannot be achieved. Humans are still an important part of HCI systems, and intelligent systems should be able to “feel” and “predict” human intentions in order to achieve dynamic coordination between humans and machines. Intent recognition is very important to improve the accuracy and efficiency of the HCI system. However, it is far from enough to focus only on explicit intent. There is a lot of vague and hidden implicit intent in the process of human–computer interaction. Based on passive brain–computer interface (pBCI) technology, this paper proposes a method to integrate humans into HCI systems naturally, which is to establish an intent-based HCI model and automatically recognize the implicit intent according to human EEG signals. In view of the existing problems of few divisible patterns and low efficiency of implicit intent recognition, this paper finally proves that EEG can be used as the basis for judging human implicit intent through extracting multi-task intention, carrying out experiments, and constructing algorithmic models. The CSP + SVM algorithm model can effectively improve the EEG decoding performance of implicit intent in HCI, and the effectiveness of the CSP algorithm on intention feature extraction is further verified by combining 3D space visualization. The translation of implicit intent information is of significance for the study of intent-based HCI models, the development of HCI systems, and the improvement of human–machine collaboration efficiency.

Towards Voice Reconstruction from EEG during Imagined Speech

Translating imagined speech from human brain activity into voice is a challenging and absorbing research issue that can provide new means of human communication via brain signals. Efforts to reconstruct speech from brain activity have shown their potential using invasive measures of spoken speech data, but have faced challenges in reconstructing imagined speech. In this paper, we propose NeuroTalk, which converts non-invasive brain signals of imagined speech into the user's own voice. Our model was trained with spoken speech EEG which was generalized to adapt to the domain of imagined speech, thus allowing natural correspondence between the imagined speech and the voice as a ground truth. In our framework, an automatic speech recognition decoder contributed to decomposing the phonemes of the generated speech, demonstrating the potential of voice reconstruction from unseen words. Our results imply the potential of speech synthesis from human EEG signals, not only from spoken speech but also from the brain signals of imagined speech.

Scale-Free Dynamics in Instantaneous Alpha Frequency Fluctuations: Validation, Test-Retest Reliability and Its Relationship with Task Manipulations.

Previous studies showed that scale-free structures and long-range temporal correlations are ubiquitous in physiological signals (e.g., electroencephalography). This is supposed to be associated with optimized information processing in human brain. The instantaneous alpha frequency (IAF) (i.e., the instantaneous frequency of alpha band of human EEG signals) may dictate the resolution at which information is sampled and/or processed by cortical neurons. To the best of our knowledge, no research has examined the scale-free dynamics and potential functional significance of IAF. Here, through three studies (Study 1: 25 participants; Study 2: 82 participants; Study 3: 26 participants), we investigated the possibility that time series of IAF exhibit scale-free property through maximum likelihood based detrended fluctuation analysis (ML-DFA). This technique could provide the scaling exponent (i.e., DFA exponent) on the basis of presence of scale-freeness being validated. Then the test-retest reliability (Study 1) and potential influencing factors (Study 2 and Study 3) of DFA exponent of IAF fluctuations were investigated. Firstly, the scale-free property was found to be inherent in IAF fluctuations with fairly high test-retest reliability over the parietal-occipital region. Moreover, the task manipulations could potentially modulate the DFA exponent of IAF fluctuations. Specifically, in Study 2, we found that the DFA exponent of IAF fluctuations in eye-closed resting-state condition was significantly larger than that in eye-open resting-state condition. In Study 3, we found that the DFA exponent of IAF fluctuations in eye-open resting-state condition was significantly larger than that in visual n-back tasks. The DFA exponent of IAF fluctuations in the 0-back task was significantly larger than in the 2-back and 3-back tasks. The results in studies 2 and 3 indicated that: (1) a smaller DFA exponent of IAF fluctuations should signify more efficient online visual information processing; (2) the scaling property of IAF fluctuations could reflect the physiological arousal level of participants.

Fatigue Detection of XYZ Drivers based on Human Brain Wave EEG Signals

The cause of death due to traffic accidents isnow increasingly common. One of the main factors causing thisaccident is driver fatigue. This can happen because the driveris not aware of his tired mental state. Of course mental fatiguecan cause a lack of concentration while driving. This mentalfatigue can be detected by analyzing the brain waves throughthe EEG signal from the driver. This brain wave analysis canbe done by various methods. In this study, the authorsconducted a brain wave-based detection of mental fatigueusing the Fourier transform and Support Vector Machine. TheEEG signal data will be feature extracted using the FourierTransform. Then, the results of this extraction will be used forthe classification process with the Support Vector Machinemethod. Based on the experimental results, the classification ofmental fatigue using a Support Vector Machine with a linearkernel obtained an average accuracy of 85%.

EEG-based spatial elements optimisation design method

In the field of digital design, a recent hot topic is the study of the interaction between spatial environment design and human factors. Electroencephalogram (EEG) and eye tracking can be used as quantitative analysis methods for architectural space evaluation; however, conclusions from existing studies on improving the quality of spatial environments based on human factors tend to remain qualitative. In order to realise the quantitative optimisation design of spatial elements from human physiological data, this research used the digital space optimisation method and perceptual evaluation research. In this way, it established an optimisation method for built space elements in real-time using human psychological indicators. Firstly, this method used the specific indicators of the Meditation value and Attention value in the human EEG signal, taking the ThinkGear AM (TGAM) module as the optimisation objective, the architectural space colour and the window size as the optimisation object, and the multi-objective genetic algorithm as the optimisation tool. Secondly, this research combined virtual reality scenarios and parametric linkage models to realise this optimisation method to establish a tool platform and workflow. Thirdly, this study took the optimisation of a typical living space as an example and recruited 50 volunteers to participate in an optimisation experiment. The results indicated that with the iterative optimisation of the multi-objective genetic algorithm, the specific EEG index decreases significantly and the standard deviation of the in-dex fluctuates and decreases during the iterative process, which further indicates that the optimisation method established in this study with the specific EEG index as the optimisation objective is effective and feasible. In addition, this study laid the foundation for more EEG indicators and more complex spatial element opti-misation research in the future.

Multi-Modal Emotion Recognition Combining Face Image and EEG Signal

Face expression can be used to identify human emotions, but it is easy to misjudge when hidden artificially. In addition, the sentiment recognition of a single mode often results in low recognition rate due to the characteristics of the single mode itself. In order to solve the mentioned problems, the spatio-temporal neural network and the separable residual network proposed by fusion can realize the emotion recognition of EEG and face. The average recognition rates of EEG and face data sets are 78.14% and 70.89%, respectively, and the recognition rates of decision fusion on DEAP data sets are 84.53%. Experimental results show that compared with the single mode, the proposed two-mode emotion recognition architecture has better performance, and can well integrate the emotional information contained in human face visual signals and EEG signals.

Decomposing the role of alpha oscillations during brain maturation.

Childhood and adolescence are critical stages of the human lifespan, in which fundamental neural reorganizational processes take place. A substantial body of literature investigated accompanying neurophysiological changes, focusing on the most dominant feature of the human EEG signal: the alpha oscillation. Recent developments in EEG signal-processing show that conventional measures of alpha power are confounded by various factors and need to be decomposed into periodic and aperiodic components, which represent distinct underlying brain mechanisms. It is therefore unclear how each part of the signal changes during brain maturation. Using multivariate Bayesian generalized linear models, we examined aperiodic and periodic parameters of alpha activity in the largest openly available pediatric dataset (N=2529, age 5-22 years) and replicated these findings in a preregistered analysis of an independent validation sample (N=369, age 6-22 years). First, the welldocumented age-related decrease in total alpha power was replicated. However, when controlling for the aperiodic signal component, our findings provided strong evidence for an age-related increase in the aperiodic-adjusted alpha power. As reported in previous studies, also relative alpha power revealed a maturational increase, yet indicating an underestimation of the underlying relationship between periodic alpha power and brain maturation. The aperiodic intercept and slope decreased with increasing age and were highly correlated with total alpha power. Consequently, earlier interpretations on age-related changes of total alpha power need to be reconsidered, as elimination of active synapses rather links to decreases in the aperiodic intercept. Instead, analyses of diffusion tensor imaging data indicate that the maturational increase in aperiodic-adjusted alpha power is related to increased thalamocortical connectivity. Functionally, our results suggest that increased thalamic control of cortical alpha power is linked to improved attentional performance during brain maturation.

Survey of emotion recognition methods using EEG information

Emotion is an indispensable part of human emotion, which affects human normal physiological activities and daily life decisions. Human emotion recognition is a critical technology in artificial intelligence, human-computer interaction, and other fields. The brain is the information processing and control center of the human body. Electroencephalogram (EEG) physiological signals are generated directly by the central nervous system, closely related to human emotions. Therefore, EEG signals can objectively and now reflect the human emotional state in real-time. In recent years, with the development of the brain-computer interface, the acquisition and analysis technology of human EEG signals has become increasingly mature, so more and more researchers use the research method based on EEG signals to study emotion recognition. EEG processing plays a vital role in emotion recognition. This paper presents a recent research report on emotion recognition. This paper introduces the related analysis methods and research contents from the aspects of emotion induction, EEG preprocessing, feature extraction, and emotion classification and compares the advantages and disadvantages of these methods. This paper summarizes the problems existing in current research methods. This paper discusses the research direction of emotion classification based on EEG information.

The search for statistical patterns of pathological activity in human EEG signals in focal epilepsy

Modern data science faces a lot of challenges, one of which is the search for diagnostic criteria for neurological diseases. New methods of statistical analysis are actively applied in the field of biophysics to solve this issue. In this paper we apply the Memory Functions Formalism to analyze electroencephalogram signal recordings in the sleeping state of 8 healthy subjects and 19 patients with nocturnal lobe epilepsy. We observe the considerable difference of statistical memory effects and fractal properties at the pathology in comparison with the control group. Furthermore, we reveal significant alterations in brain rhythms at power spectra of statistical memory functions for two groups of subjects. As a result, we show that the application of the statistical analysis methodology of bioelectrical brain cortex activity recordings, after appropriate verification, can be useful in the search for diagnostic criteria of nocturnal frontal lobe epilepsy.

Gated Transformer for Decoding Human Brain EEG Signals.

In this work, we propose to use a deep learning framework for decoding the electroencephalogram (EEG) signals of human brain activities. More specifically, we learn an end-to-end model that recognizes natural images or motor imagery by the EEG data that is collected from the corresponding human neural activities. In order to capture the temporal information encoded in the long EEG sequences, we first employ an enhanced version of Transformer, i.e., gated Transformer, on EEG signals to learn the feature representation along a sequence of embeddings. Then a fully-connected Softmax layer is used to predict the classification results of the decoded representations. To demonstrate the effectiveness of the gated Transformer approach, we conduct experiments on the image classification task for a human brain-visual dataset and the classification task for a motor imagery dataset. The experimental results show that our method achieves new state-of-the-art performance compared to multiple existing methods that are widely used for EEG classification.

Human EEG and Recurrent Neural Networks Exhibit Common Temporal Dynamics During Speech Recognition.

Recent deep-learning artificial neural networks have shown remarkable success in recognizing natural human speech, however the reasons for their success are not entirely understood. Success of these methods might be because state-of-the-art networks use recurrent layers or dilated convolutional layers that enable the network to use a time-dependent feature space. The importance of time-dependent features in human cortical mechanisms of speech perception, measured by electroencephalography (EEG) and magnetoencephalography (MEG), have also been of particular recent interest. It is possible that recurrent neural networks (RNNs) achieve their success by emulating aspects of cortical dynamics, albeit through very different computational mechanisms. In that case, we should observe commonalities in the temporal dynamics of deep-learning models, particularly in recurrent layers, and brain electrical activity (EEG) during speech perception. We explored this prediction by presenting the same sentences to both human listeners and the Deep Speech RNN and considered the temporal dynamics of the EEG and RNN units for identical sentences. We tested whether the recently discovered phenomenon of envelope phase tracking in the human EEG is also evident in RNN hidden layers. We furthermore predicted that the clustering of dissimilarity between model representations of pairs of stimuli would be similar in both RNN and EEG dynamics. We found that the dynamics of both the recurrent layer of the network and human EEG signals exhibit envelope phase tracking with similar time lags. We also computed the representational distance matrices (RDMs) of brain and network responses to speech stimuli. The model RDMs became more similar to the brain RDM when going from early network layers to later ones, and eventually peaked at the recurrent layer. These results suggest that the Deep Speech RNN captures a representation of temporal features of speech in a manner similar to human brain.

Capturing the Development of Internal Representations in a High-Performing Deep Network for Sleep Stage Classification

Deep neural networks can provide accurate automated classification of human sleep signals into sleep stages that enables more effective diagnosis and treatment of sleep disorders. We develop a deep convolutional neural network (CNN) that attains state-of-the-art sleep stage classification performance on input data consisting of human sleep EEG and EOG signals. Nested cross-validation is used for optimal model selection and reliable estimation of out-of-sample classification performance. The resulting network attains a classification accuracy of 84.50 pm 0.13%; its performance exceeds human expert inter-scorer agreement, even on single-channel EEG input data, therefore providing more objective and consistent labeling than human experts demonstrate as a group. We focus on analyzing the learned internal data representations of our network, with the aim of understanding the development of class differentiation ability across the layers of processing units, as a function of layer depth. We approach this problem visually, using t-Stochastic Neighbor Embedding (t-SNE), and propose a pooling variant of Centered Kernel Alignment (CKA) that provides an objective quantitative measure of the development of sleep stage specialization and differentiation with layer depth. The results reveal a monotonic progression of both of these sleep stage modeling abilities as layer depth increases.

Clinical Diagnosis of Psychiatry Based on Electroencephalography

At present, neurophysiological signals used for neuro feedback are EEG (Electroencephalogram), functional magnetic resonance imaging. Among them, the acquisition of EEG signals has the advantages of non-invasive way with low cost. It has been widely used in brain-machine interface technology in recent years. Important progress has been made in rehabilitation and environmental control. However, neural feedback and brainmachine interface technology are completely similar in signal acquisition, signal feature extraction, and pattern classification. Therefore, the related research results of brain-machine interface can be used to closely cooperate with clinical needs to research and develop neural feedback technology based on EEG. Based on neurophysiology and brain-machine interface technology, this paper develops a neural feedback training system based on the acquisition and analysis of human EEG signals. Aiming at the autonomous rhythm components in the EEG signal, such as sensorimotor rhythm and alpha rhythm, the characteristic parameters are extracted through real-time EEG signal processing to generate feedback information, and the subject is self-regulated and trained from a physiological-psychological perspective by providing adjuvant treatment, a practical and stable treatment platform for the clinic.

Steady-State Visually Evoked Potentials and Feature-based Attention: Preregistered Null Results and a Focused Review of Methodological Considerations.

Feature-based attention is the ability to selectively attend to a particular feature (e.g., attend to red but not green items while looking for the ketchup bottle in your refrigerator), and steady-state visually evoked potentials (SSVEPs) measured from the human EEG signal have been used to track the neural deployment of feature-based attention. Although many published studies suggest that we can use trial-by-trial cues to enhance relevant feature information (i.e., greater SSVEP response to the cued color), there is ongoing debate about whether participants may likewise use trial-by-trial cues to voluntarily ignore a particular feature. Here, we report the results of a preregistered study in which participants either were cued to attend or to ignore a color. Counter to prior work, we found no attention-related modulation of the SSVEP response in either cue condition. However, positive control analyses revealed that participants paid some degree of attention to the cued color (i.e., we observed a greater P300 component to targets in the attended vs. the unattended color). In light of these unexpected null results, we conducted a focused review of methodological considerations for studies of feature-based attention using SSVEPs. In the review, we quantify potentially important stimulus parameters that have been used in the past (e.g., stimulation frequency, trial counts) and we discuss the potential importance of these and other task factors (e.g., feature-based priming) for SSVEP studies.

Deep Learning for Automated Feature Discovery and Classification of Sleep Stages.

Convolutional neural networks (CNN) have demonstrated state-of-the-art classification results in image categorization, but have received comparatively little attention for classification of one-dimensional physiological signals. We design a deep CNN architecture for automated sleep stage classiffication of human sleep EEG and EOG signals. The CNN proposed in this paper amply outperforms recent work that uses a different CNN architecture over a single-EEG-channel version of the same dataset. We show that the performance gains achieved by our network rely mainly on network depth, and not on the use of several signal channels. Performance of our approach is on par with human expert inter-scorer agreement. By examining the internal activation levels of our CNN, we find that it spontaneously discovers signal features such as sleep spindles and slow waves that figure prominently in sleep stage categorization as performed by human experts.

Anticipated synchronization in human EEG data: Unidirectional causality with negative phase lag.

Understanding the functional connectivity of the brain has become a major goal of neuroscience. In many situations the relative phase difference, together with coherence patterns, has been employed to infer the direction of the information flow. However, it has been recently shown in local field potential data from monkeys the existence of a synchronized regime in which unidirectionally coupled areas can present both positive and negative phase differences. During the counterintuitive regime, called anticipated synchronization (AS), the phase difference does not reflect the causality. Here we investigate coherence and causality at the alpha frequency band (f∼10Hz) between pairs of electroencephalogram (EEG) electrodes in humans during a GO/NO-GO task. We show that human EEG signals can exhibit anticipated synchronization, which is characterized by a unidirectional influence from an electrode A to an electrode B, but the electrode B leads the electrode A in time. To the best of our knowledge, this is the first verification of AS in EEG signals and in the human brain. The usual delayed synchronization (DS) regime is also present between many pairs. DS is characterized by a unidirectional influence from an electrode A to an electrode B and a positive phase difference between A and B which indicates that the electrode A leads the electrode B in time. Moreover we show that EEG signals exhibit diversity in the phase relations: the pairs of electrodes can present in-phase, antiphase, or out-of-phase synchronization with a similar distribution of positive and negative phase differences.

EEG Research Based on the Influence of Different Music Effects

Music has become one of the methods for regulating emotions in the daily life of most people. Different concerts bring different emotional changes to people, but the effect of music on the changes of human brain and EEG signals is little known. This paper uses the brain wave (EEG) experiment to study the effect of different musical stimuli on attention, and focuses on comparing the effects of brain wave music and heavy metal music that are widely studied on attention. In this paper, an experiment was designed to collect the EEG signals of 15 students in three states: brainwave music (slow 80BPM), heavy metal music (fast 140BPM) and resting EEG signals. Independent component analysis (ICA) preprocesses the EEG signal, and uses T, T, and T to analyze the characteristics of the α, β and θ wave signals obtained by wavelet transform. Through the experimental analysis of this article, it is found that brain wave music has a significant effect on the alpha wave (P<0.05), and heavy metal concerts have an effect on the power value of the beta wave. The experimental results show that brain wave music, which is the slow board music, can better pay attention. The concentration of force, and heavy metal music (clapper music) will make the attention more distracting.