EEG Signals Research Articles

Domain Adaptation for Fear of Heights Classification in a VR Environment Based on EEG and ECG

AbstractThree levels of fear of heights were detected in subjects with different severities of acrophobia, based on the electroencephalographic (EEG) and electrocardiographic (ECG) signals. The study aims to demonstrate the feasibility of a data-fusion-based method for real-time assessment of the fear of heights intensity to integrate into adaptive Virtual Reality Exposure Therapy for acrophobia. The generalization performance of classification tasks on fear states is improved by exploiting both trait-based clustering and Domain Adaptation methods. Participants were gradually exposed to increasing height levels through a Virtual Reality (VR) scenario representing a canyon. The initial severity of fear of heights, the level of distress at each height, and the anxiety level before and after the exposure were assessed through the Acrophobia Questionnaire, the Subjective Unit of Distress, and the State and Trait Anxiety Inventory, respectively. The Simulator Sickness Questionnaire was administered to exclude possible motion sickness interference in the experiment. The EEG and ECG signals were acquired through a 32-channel headset and 1 Lead ECG derivation during the exposure to the eliciting VR scenario. Four classifiers (i.e. Support Vector Machines, Deep Neural Networks, Random Forests, and k-Nearest Neighbors) were adopted in the experimental environment. Preliminary tests were performed in a within-subject experiment, achieving the best classification accuracy of $$87.1 \% \pm 7.8 \%$$ 87.1 % ± 7.8 % with a Deep Neural Network. As the cross-subject approach is concerned, three strategies, namely Domain Adaptation (DA), data fusion (combining EEG with ECG), and participant clustering (based on the acrophobia severity), were evaluated. DA resulted in the most effective strategies by determining an improvement of more than 20 % in classification accuracy. Random Forest performed the best classification accuracy for the severe acrophobia cluster with a mean of $$63.6 \%$$ 63.6 % and a standard deviation of $$ 13.4 \%$$ 13.4 % over three classes by exploiting Stratified Normalization.

A Data Augmentation Method for Motor Imagery EEG Signals Based on DCGAN-GP Network.

Motor imagery electroencephalography (EEG) signals have garnered attention in brain-computer interface (BCI) research due to their potential in promoting motor rehabilitation and control. However, the limited availability of labeled data poses challenges for training robust classifiers. In this study, we propose a novel data augmentation method utilizing an improved Deep Convolutional Generative Adversarial Network with Gradient Penalty (DCGAN-GP) to address this issue. We transformed raw EEG signals into two-dimensional time-frequency maps and employed a DCGAN-GP network to generate synthetic time-frequency representations resembling real data. Validation experiments were conducted on the BCI IV 2b dataset, comparing the performance of classifiers trained with augmented and unaugmented data. Results demonstrated that classifiers trained with synthetic data exhibit enhanced robustness across multiple subjects and achieve higher classification accuracy. Our findings highlight the effectiveness of utilizing a DCGAN-GP-generated synthetic EEG data to improve classifier performance in distinguishing different motor imagery tasks. Thus, the proposed data augmentation method based on a DCGAN-GP offers a promising avenue for enhancing BCI system performance, overcoming data scarcity challenges, and bolstering classifier robustness, thereby providing substantial support for the broader adoption of BCI technology in real-world applications.

Compensatory brain activity pattern is not present in older adults during the n-back task performance-Findings based on EEG frequency analysis.

Cognitive ability is one of the most important enablers for successful aging. At the same time, cognitive decline is a well-documented phenomenon accompanying the aging process. Nevertheless, it is acknowledged that aging can also be related to positive processes that allow one to compensate for the decline. These processes include the compensatory brain activity of older adults primarily investigated using fMRI and PET. To strengthen the cognitive interpretation of compensatory brain activity in older adults, we searched for its indicators in brain activity measured by EEG. The study sample comprised 110 volunteers, including 50 older adults (60-75 years old) and 60 young adults (20-35 years old) who performed 1-back, 2-back, and 3-back tasks while recording the EEG signal. The study analyzed (1) the level of cognitive performance, including sensitivity index, the percentage of correct answers to the target, and the percentage of false alarm errors; (2) theta and alpha power for electrodes located in the frontal-midline (Fz, AF3, AF4, F3, F4, FC1, and FC2) and the centro-parietal (CP1, CP2, P3, P4, and Pz) areas. Cognitive performance was worse in older adults than in young adults, which manifested in a significantly lower sensitivity index and a significantly higher false alarm error rate at all levels of the n-back task difficulty. Simultaneously, performance worsened with increasing task difficulty regardless of age. Significantly lower theta power in the older participants was observed at all difficulty levels, even at the lowest one, where compensatory activity was expected. At the same time, at this difficulty level, cognitive performance was worse in older adults than in young adults, which could reduce the chances of observing compensatory brain activity. The significant decrease in theta power observed in both age groups with rising task difficulty can reflect a declining capacity for efficient cognitive functioning under increasing demands rather than adapting to this increase. Moreover, in young adults, alpha power decreased to some extent with increasing cognitive demand, reflecting adaptation to them, while in older adults, no analogous pattern was observed. In conclusion, based on the results of the current study, the presence of compensatory activity in older adults cannot be inferred.

Seizure detection via deterministic learning feature extraction

Epileptic seizures have a significant impact on the well-being of a large number of individuals worldwide. Utilizing electroencephalographic (EEG) signals for automatic seizure detection proves to be a valuable solution. However, dealing with raw EEG signals is inherently complex, necessitating a preliminary step of feature extraction prior to detection. Traditional feature extraction methods often amalgamate various types of features for seizure detection, as each type typically captures specific properties. In contrast, this paper focuses on detecting seizures by analyzing the system dynamics. The proposed Deterministic Learning Feature Extraction (DLFE) method extracts a single type of nonlinear dynamical feature rooted in the EEG system dynamics. DLFE employs deterministic learning to discern the inherent system dynamics of the EEG under both seizure and normal conditions. Through the feature extraction process, the infinite-dimensional system dynamics are transformed into feature vectors, exhibiting distinct distributions in seizure and normal states. This disparity can be effectively utilized for classification using standard classifiers. The performance of the proposed seizure detection method was assessed using the CHB-MIT and Bonn datasets. The average classification accuracy was found to be 98.63% with a specificity of 99.19% and a sensitivity of 98.06% on CHB-MIT dataset. Compared with the latest similar methods, the accuracy, specificity and sensitivity are improved by 0.31%, 0.21% and 0.05% respectively. Moreover, the performance was achieved with the short-time interval EEG signals within a few channels. The average classification accuracy was found to be 99.90% with a 0.22% improvement on Bonn dataset, which indicates the good generalization performance.

Gabor filter-based statistical features for ADHD detection.

Attention deficit/hyperactivity disorder (ADHD) is a neuropsychological disorder that occurs in children and is characterized by inattention, impulsivity, and hyperactivity. Early and accurate diagnosis of ADHD is very important for effective intervention. The aim of this study is to develop a computer-aided approach to detecting ADHD using electroencephalogram (EEG) signals. Specifically, we explore a Gabor filter-based statistical features approach for the classification of EEG signals into ADHD and healthy control (HC). The EEG signal is processed by a bank of Gabor filters to obtain narrow-band signals. Subsequently, a set of statistical features is extracted. The computed features are then subjected to feature selection. Finally, the obtained feature vector is given to a classifier to detect ADHD and HC. Our approach achieves the highest classification accuracy of 96.4% on a publicly available dataset. Furthermore, our approach demonstrates better classification accuracy than the existing methods.

Electroencephalographic Signal Data Augmentation Based on Improved Generative Adversarial Network.

EEG signals combined with deep learning play an important role in the study of human-computer interaction. However, the limited dataset makes it challenging to study EEG signals using deep learning methods. Inspired by the GAN network in image generation, this paper presents an improved generative adversarial network model L-C-WGAN-GP to generate artificial EEG data to augment training sets and improve the application of BCI in various fields. The generator consists of a long short-term memory (LSTM) network and the discriminator consists of a convolutional neural network (CNN) which uses the gradient penalty-based Wasserstein distance as the loss function in model training. The model can learn the statistical features of EEG signals and generate EEG data that approximate real samples. In addition, the performance of the compressed sensing reconstruction model can be improved by using augmented datasets. Experiments show that, compared with the existing advanced data amplification techniques, the proposed model produces EEG signals closer to the real EEG signals as measured by RMSE, FD and WTD indicators. In addition, in the compressed reconstruction of EEG signals, adding the new data reduces the loss by about 15% compared with the original data, which greatly improves the reconstruction accuracy of the EEG signals' compressed sensing.

Intelligent learners distraction and drowsiness prediction through EEG signal and iris angel position with brain vision algorithm

Distraction and fatigue are serious issues in online learning, and they directly impact educational outcomes. To achieve excellent academic achievement, students need to focus on their studies without being distracted or fatigued. Learners frequently overlook crucial information, directions, and concepts while they are passive and sleepy. They tend to miss important content, instructions, and concepts. Iris Angle Position (IAP) and electroencephalography (EEG) were used in this model to identify the behaviour of learners. Specifically, a Deep Convolutional Neural Network (DCNN) is constructed to extract IAP in order to accurately capture the learner’s facial area. EEG signals are effectively handled and sorted using deep reinforcement learning (DRL). The learners’ facial landmarks are retrieved from a frame using the dlib toolbox. Only eye landmark points from face landmarks alone are focused on in order to determine the learner’s behaviour. When the learners EEG signals and Iris positions are monitored simultaneously, it’s helpful to identify the learner’s fatigue state (LFS) and the learner’s distraction state (LDS). The Brain Vision Algorithm (BVA) uses iris position and minimal facial landmarks, along with brain activity, to properly identify the learner’s level of distraction and exhaustion. When a student is detected as being preoccupied or sleepy, an alert goes off automatically, and the educator gets performance feedback. Iris position data and brain-computer interface-based EEG signal values are utilised to identify distraction and sleepiness. Comparative tests have demonstrated that this innovative method offers fast and high-accuracy student activity detection in virtual learning settings. Applying the suggested approach to different existing classifiers yields an F-Score of 91.92%, a recall of 93.87%, and a precision of 92.37% . The results showed that the detection rates for both distracted and sleepy phases were higher than those attained with other currently used techniques.

Blocked versus interleaved: How range contexts modulate time perception and its EEG signatures.

Accurate time perception is a crucial element in a wide range of cognitive tasks, including decision-making, memory, and motor control. One commonly observed phenomenon is that when given a range of time intervals to consider, people's estimates often cluster around the midpoint of those intervals. Previous studies have suggested that the range of these intervals can also influence our judgments, but the neural mechanisms behind this "range effect" are not yet understood. We used both behavioral tests and electroencephalographic (EEG) measures to understand how the range of sample time intervals affects the accuracy of people's subsequent time estimates. Study participants were exposed to two different setups: In the "blocked-range" (BR) session, short and long intervals were presented in separate blocks, whereas in the "interleaved-range" (IR) session, intervals of various lengths were presented randomly. Our findings indicated that the BR context led to more accurate time estimates compared to the IR context. In terms of EEG data, the BR context resulted in quicker buildup of contingent negative variation (CNV), which also reached higher amplitude levels and dissolved more rapidly during the encoding stage. We also observed an enhanced amplitude in the offset P2 component of the EEG signal. Overall, our results suggest that the variability in time intervals, as defined by their range, influences the neural processes that underlie time estimation.

Emerging Trends in EEG Signal Processing: A Systematic Review

Emerging Trends in EEG Signal Processing: A Systematic Review

EEG Microstate Associated with Trait Nostalgia.

Nostalgia, a self-related emotion characterized by its bittersweet yet predominantly positive nature, plays a vital role in shaping individual psychology and behavior. This includes impacts on mental and physical health, behavioral patterns, and cognitive functions. However, higher levels of trait nostalgia may be linked to potential adverse outcomes, such as increased loneliness, heightened neuroticism, and more intense experiences of grief. The specific electroencephalography (EEG) feature associated with individuals exhibiting trait nostalgia, and how it differs from others, remains an area of uncertainty. To address this, our study employs microstate analysis to investigate the differences in resting-state EEG between individuals with varying levels of trait nostalgia. We assessed trait nostalgia in 63 participants using the Personal Inventory of Nostalgia and collected their resting-state EEG signals with eyes closed. The results of the regression analysis indicate a significant correlation between trait nostalgia and the temporal characteristics of microstates A, B, and C. Further, the occurrence of microstate B was significantly more frequent in the high trait nostalgia group than in the low trait nostalgia group. Independent samples t-test results showed that the transition probability between microstates A and B was significantly higher in the high trait nostalgia group. These results support the hypothesis that trait nostalgia is reflected in the resting state brain activity. Furthermore, they reveal a deeper sensory immersion in nostalgia experiences among individuals with high levels of trait nostalgia, and highlight the critical role of self-referential and autobiographical memory processes in nostalgia.

Scented Solutions: Examining the Efficacy of Scent Interventions in Mitigating Driving Fatigue.

Fatigued driving threatens the safety of people's lives and property. Scent countermeasures offer minimal disruption and high efficacy, making them a promising approach. The aim of this study was to explore the application of scent countermeasures in alleviating fatigued driving. This study explored changes in EEG frequency bands (alpha, beta, theta, and gamma) and the activity of EEG metrics (R(α/β), Rθ/(α+β) and R(α+θ)/(α+β)) in the temporal lobe during driving tasks, selected fatigued driving identifiers, and aided validation by investigating subjective fatigue with the Karolinska Sleepiness Scale (KSS). The EEG indicators all increased, with a significant increase in R(α/β). R(α/β) was combined with the KSS to explore the effects of three scents, peppermint, grapefruit, and lavender, on driving fatigue. The subjective questionnaire results indicated that all three scents significantly improved driving fatigue, with significantly lower levels of driving fatigue compared to the control group. The analysis of EEG signals revealed a significant decrease in R(α/β) after the implementation of scent countermeasures. Moreover, R(α/β) was found to be lower in all three odor intervention groups compared to the control group. All three scents were found to significantly alleviate driving fatigue. The grapefruit scent had a better timely effect in relieving driving fatigue and the lavender scent had a longer effectiveness. This study provides further exploration for the application of odor interventions to alleviate driving fatigue. This study provides a practical reference for drivers to use odors to avoid fatigue in order to improve road safety.

EEG emotion recognition based on differential entropy feature matrix through 2D-CNN-LSTM network

Emotion recognition research has attracted great interest in various research fields, and electroencephalography (EEG) is considered a promising tool for extracting emotion-related information. However, traditional EEG-based emotion recognition methods ignore the spatial correlation between electrodes. To address this problem, this paper proposes an EEG-based emotion recognition method combining differential entropy feature matrix (DEFM) and 2D-CNN-LSTM. In this work, first, the one-dimensional EEG vector sequence is converted into a two-dimensional grid matrix sequence, which corresponds to the distribution of brain regions of the EEG electrode positions, and can better characterize the spatial correlation between the EEG signals of multiple adjacent electrodes. Then, the EEG signal is divided into equal time windows, and the differential entropy (DE) of each electrode in this time window is calculated, it is combined with a two-dimensional grid matrix and differential entropy to obtain a new data representation that can capture the spatiotemporal correlation of the EEG signal, which is called DEFM. Secondly, we use 2D-CNN-LSTM to accurately identify the emotional categories contained in the EEG signals and finally classify them through the fully connected layer. Experiments are conducted on the widely used DEAP dataset. Experimental results show that the method achieves an average classification accuracy of 91.92% and 92.31% for valence and arousal, respectively. The method performs outstandingly in emotion recognition. This method effectively combines the temporal and spatial correlation of EEG signals, improves the accuracy and robustness of EEG emotion recognition, and has broad application prospects in the field of emotion classification and recognition based on EEG signals.

Low-quality Video Target Detection Based on EEG Signal using Eye Movement Alignment

Low-quality Video Target Detection Based on EEG Signal using Eye Movement Alignment

Analysis of reading-task-based brain connectivity in dyslexic children using EEG signals.

Developmental dyslexia, a neurodevelopment reading disorder, can impact even children with average intelligence. The present study examined the brain connectivity in dyslexic and control children during the reading task using graph theory. 19-channel electroencephalogram (EEG) signals were recorded from 15 dyslexic children and 15 control children. Functional connectivity was estimated by measuring the EEG coherence at 19 electrode locations, and graph measures were calculated using the graph theory method. Reading task results identified deprived task performance in dyslexic children against controls. Graph measures revealed longer path length, reduced clustering coefficient and reduced network efficiencies (in theta and alpha bands) of dyslexic group. At the nodal level, we found a significant increase in delta strength (T4 and T5 electrode locations) and reduced strength in theta (T6, P4, Fp1, F8 and F3) and alpha bands (T4, T3, P4 and F3) during the reading task in dyslexic group. In conclusion, the present study identified distinct graph measures between groups when performing a reading task and showed possible evidence for compromised brain networks in dyslexic group.

Real-time Single-Channel EOG removal based on Empirical Mode Decomposition

In recent years, single-channel physiological recordings have gained popularity in portable health devices and research settings due to their convenience. However, the presence of electrooculogram (EOG) artifacts can significantly degrade the quality of the recorded data, impacting the accuracy of essential signal features. Consequently, artifact removal from physiological signals is a crucial step in signal processing pipelines. Current techniques often employ Independent Component Analysis (ICA) to efficiently separate signal and artifact sources in multichannel recordings. However, limitations arise when dealing with single or a few channel measurements in minimal instrumentation or portable devices, restricting the utility of ICA. To address this challenge, this paper introduces an innovative artifact removal algorithm utilizing enhanced empirical mode decomposition to extract the intrinsic mode functions (IMFs). Subsequently, the algorithm targets the removal of segments related to EOG by isolating them within these IMFs. The proposed method is compared with existing single-channel EEG artifact removal algorithms, demonstrating superior performance. The findings demonstrate the effectiveness of our approach in isolating artifact components, resulting in a reconstructed signal characterized by a strong correlation and a power spectrum closely resembling the ground-truth EEG signal. This outperforms the existing methods in terms of artifact removal. Additionally, the proposed algorithm exhibits significantly reduced execution time, enabling real-time online analysis.

Improving emotion perception in cochlear implant users: insights from machine learning analysis of EEG signals.

Although cochlear implants can restore auditory inputs to deafferented auditory cortices, the quality of the sound signal transmitted to the brain is severely degraded, limiting functional outcomes in terms of speech perception and emotion perception. The latter deficit negatively impacts cochlear implant users' social integration and quality of life; however, emotion perception is not currently part of rehabilitation. Developing rehabilitation programs incorporating emotional cognition requires a deeper understanding of cochlear implant users' residual emotion perception abilities. To identify the neural underpinnings of these residual abilities, we investigated whether machine learning techniques could be used to identify emotion-specific patterns of neural activity in cochlear implant users. Using existing electroencephalography data from 22 cochlear implant users, we employed a random forest classifier to establish if we could model and subsequently predict from participants' brain responses the auditory emotions (vocal and musical) presented to them. Our findings suggest thatconsistent emotion-specific biomarkers exist in cochlear implant users, which could be used to develop effective rehabilitation programs incorporating emotion perception training. This study highlights the potential of machine learning techniques to improve outcomes for cochlear implant users, particularly in terms of emotion perception.

The Neural Correlates of Food Preference among Music Kinds.

The calorie and taste choices of food have been shown to be related to the external environment, including music. Previous studies have mostly focused on manipulating basic auditory parameters, with few scholars exploring the impact of complex musical parameters on food selection. This study explored the effects of different kinds of music (classical, rock, jazz, and hip-hop) on food liking based on the calories (high and low) and taste (sweet and salty) using event-related potentials (ERPs). Twenty-four participants (8 males, 16 females) were recruited from Southwest University, China to participate in the food liking task using a Likert seven-point rating and simultaneously recording EEG signals (N2, P2, N3, and LPC). This study used repeated-measures analyses of covariances and found that the score of the high-calorie foods was greater than that of the low-calorie foods. Additionally, results revealed that the score in classical music was greatest for sweet foods, while there was no difference among music kinds in the salty foods. The ERP results showed that P2 amplitudes were greater for sweet foods than those for the salty foods. N2 amplitudes for the salty foods were greater than those for the sweet foods during rock music; in addition, N2 amplitudes during hip-hop music were greatest for sweet foods. However, N2 amplitudes during rock music were the greatest for salty foods. The results also revealed that N2 amplitudes during hip-hop music were greater than those during jazz music. This study provides unique operational insights for businesses.

Transfer learning based epileptic seizure classification using scalogram images of EEG signals

Transfer learning based epileptic seizure classification using scalogram images of EEG signals

Quantification of EEG Characteristics for Epileptic Seizure Detection and Monitoring of Anaesthesia Using Spectral Analysis

Epilepsy is considered one of the primary neurological disorders, and its treatment requires abundant technological assistance. General Anaesthesia induces distinct patterns in brain activity, with the most common being a gradual increase in low-frequency signals as the level of Anaesthesia deepens. In this instance, a method of validating epileptic seizures and Anaesthesia through the utilization of electroencephalogram (EEG) data, acquired non-invasively, is introduced. Epileptic seizures and detection of the presence of Anaesthesia approaches make use of discrete Laplace Transformation (LT), Discrete Cosine Transformation (DCT), and Fast Fourier Transform (FFT). Here, it is discussed how power spectral analysis (PSA) helps study EEG characteristics in detecting epileptic behavior and the presence of Anaesthesia. A dataset of EEG (Epileptic and Anaesthesia), which is available publicly [1,2], has been used in the propounded technique using FIR filters and LT, DCT, and FFT are used to store and process 16 channel data. Power Spectrum Density (PSD) and its average were contrasted against a specific spectrum and frequency range of a typical EEG signal to obtain the results. This work uses a technique to determine whether the patient being studied is epileptic and awake or anesthetized.

VLSI design for adjustable compression rate in lossless/lossy compression of EEG signal

In the realm of remote health monitoring, the compression of electroencephalography (EEG) signals at the edge holds significant importance. This paper presents a VLSI structure capable of adjustable compression rates to achieve both lossless and lossy compression of EEG data, catering to environments requiring flexible adjustment of compression types and rates. The architecture integrates two-dimensional discrete wavelet transform (2D-DWT) and the Set Partitioning in Hierarchical Trees (SPIHT). Notably, the Haar wavelet lifting structure is employed in 2D-DWT, and the SPIHT scanning sequence is optimized for enhanced resource efficiency. The proposed algorithm is validated using the CHB-MIT Scalp EEG Database, demonstrating compression ratios of ≥1.95 for normal EEG and ≥1.69 for epileptic EEG. Through synthesis and layout design in the 130 nm CMOS technology, the resulting circuit achieves a maximum clock frequency of 57 MHz and occupies an area of 81 kμm2, with an average power consumption of less than 0.71nj per data for 16-bit data. These findings affirm the efficacy of the proposed VLSI structure for achieving efficient lossless/lossy compression of edge EEG data.