Electroencephalography Data Research Articles

Multiclass classification of imagined speech EEG using noise-assisted multivariate empirical mode decomposition and multireceptive field convolutional neural network.

In this study, we classified electroencephalography (EEG) data of imagined speech using signal decomposition and multireceptive convolutional neural network. The imagined speech EEG with five vowels /a/, /e/, /i/, /o/, and /u/, and mute (rest) sounds were obtained from ten study participants. First, two different signal decomposition methods were applied for comparison: noise-assisted multivariate empirical mode decomposition and wavelet packet decomposition. Six statistical features were calculated from the decomposed eight sub-frequency bands EEG. Next, all features obtained from each channel of the trial were vectorized and used as the input vector of classifiers. Lastly, EEG was classified using multireceptive field convolutional neural network and several other classifiers for comparison. We achieved an average classification rate of 73.09 and up to 80.41% in a multiclass (six classes) setup (Chance: 16.67%). In comparison with various other classifiers, significant improvements for other classifiers were achieved (p-value < 0.05). From the frequency sub-band analysis, high-frequency band regions and the lowest-frequency band region contain more information about imagined vowel EEG data. The misclassification and classification rate of each vowel imaginary EEG was analyzed through a confusion matrix. Imagined speech EEG can be classified successfully using the proposed signal decomposition method and a convolutional neural network. The proposed classification method for imagined speech EEG can contribute to developing a practical imagined speech-based brain-computer interfaces system.

Classification of Moderate and Advanced Alzheimer's Patients Using Radial Basis Function Based Neural Networks Initialized with Fuzzy Logic

BackgroundAlzheimer's disease can be diagnosed through various clinical methods. Among them, electroencephalography has proven to be a powerful, non-invasive, affordable, and painless tool for its diagnosis. ObjectivesIn this study, eight machine learning (ML) approaches, including SVM, BLDA, DT, GNB, KNN, RF, and deep learning (DL) methods such as RNN and RBF, were employed to classify Alzheimer's disease into two stages: moderate Alzheimer's disease (ADM) and advanced Alzheimer's disease (ADA). Material and methodsTo this aim, electroencephalography data collected from five different hospitals over a decade has been used. A novel method based on neural networks has been proposed to increase accuracy and obtain fast classification times. ResultsResults show that deep neuronal networks based on radial basis functions initialized with fuzzy means achieved the best balanced accuracy with 96.66% accuracy in ADA classification and 93.31% accuracy in ADM classification. ConclusionApart from improving accuracy, it is noteworthy that this algorithm had never been used before to classify patients with Alzheimer's disease.

Can we dissociate hypervigilance to social threats from altered perceptual decision-making processes in lonely individuals? An exploration with Drift Diffusion Modeling and event-related potentials.

It has been hypothesized that lonely individuals demonstrate hypervigilance toward social threats. However, recent studies have raised doubts about the reliability of tasks commonly used to measure attentional biases toward threats. Two alternative approaches have been suggested to overcome the limitations of traditional analysis of attentional bias. First, the neurophysiological indicators of orienting to threats were shown to have superior psychometric characteristics compared to overt measures of behavioral performance. The second approach involves utilizing computational modeling to isolate latent components corresponding to specific cognitive mechanisms from observable data. To test the usefulness of these approaches in loneliness research, we analyzed behavioral and electroencephalographic (EEG) data from 26 lonely and 26 non-lonely participants who performed a dot-probe task using a computational modeling approach. We applied the Drift Diffusion Model (DDM) and extracted N2pc-an event-related potential that serves as an indicator of spatial attention. No evidence for social threat hypervigilance has been found in DDM parameters nor in N2pc characteristics in the current study. However, we did observe decreased drift rate and increased variability in drift rate between trials within the lonely group, indicating reduced efficiency in perceptual decision-making among lonely individuals. These effects were not detected using standard behavioral measures used in the dot-probe paradigm. Given that DDM indicators were sensitive to differences in perceptual discrimination between the two groups, even when no overt differences were found in standard behavioral measures, it may be postulated that computational approaches offer a more comprehensive understanding of cognitive processes.

A Potential Source of Bias in Group-Level EEG Microstate Analysis.

Microstate analysis is a promising technique for analyzing high-density electroencephalographic data, but there are multiple questions about methodological best practices. Between and within individuals, microstates can differ both in terms of characteristic topographies and temporal dynamics, which leads to analytic challenges as the measurement of microstate dynamics is dependent on assumptions about their topographies. Here we focus on the analysis of group differences, using simulations seeded on real data from healthy control subjects to compare approaches that derive separate sets of maps within subgroups versus a single set of maps applied uniformly to the entire dataset. In the absence of true group differences in either microstate maps or temporal metrics, we found that using separate subgroup maps resulted in substantially inflated type I error rates. On the other hand, when groups truly differed in their microstate maps, analyses based on a single set of maps confounded topographic effects with differences in other derived metrics. We propose an approach to alleviate both classes of bias, based on a paired analysis of all subgroup maps. We illustrate the qualitative and quantitative impact of these issues in real data by comparing waking versus non-rapid eye movement sleep microstates. Overall, our results suggest that even subtle chance differences in microstate topography can have profound effects on derived microstate metrics and that future studies using microstate analysis should take steps to mitigate this large source of error.

Take your time: Slow brain rhythms predict fluid intelligence

Evidence is mixed whether fluid intelligence (Gf) is associated with increased or decreased alpha and beta band activity (7–30 Hz). Moreover, the Gf relationship with the delta and theta band activity (1–7 Hz) is unknown. We recorded electroencephalographic (EEG) data in 160 healthy adults solving Raven's Advanced Progressive Matrices with a randomized item order to control for item difficulty unaffected by sequential effects. The participants studied each matrix for 30 s before the response bank onset, so we could track the time course of neural activity during problem solving. We measured Gf using six tests. Gf positively correlated with the delta band power, while there was no correlation with the theta band power. For almost all of the participants, we identified the specific slow rhythm frequency, which varied in power as a function of item difficulty. We observed that the lower this frequency, the higher Gf, but only in men. Finally, the alpha and low-beta activity correlated negatively with Gf after we had filtered out the activity during idle intervals (the latter reflecting waiting for the response bank). Overall, the brain activity in the delta, alpha, and beta bands explained 22.6% of Gf variance.

Automated Clinical Impression Generation for Medical Signal Data Searches

Medical retrieval systems have become significantly important in clinical settings. However, commercial retrieval systems that heavily rely on term-based indexing face challenges when handling continuous medical data, such as electroencephalography data, primarily due to the high cost associated with utilizing neurologist analyses. With the increasing affordability of data recording systems, it becomes increasingly crucial to address these challenges. Traditional procedures for annotating, classifying, and interpreting medical data are costly, time consuming, and demand specialized knowledge. While cross-modal retrieval systems have been proposed to address these challenges, most concentrate on images and text, sidelining time-series medical data like electroencephalography data. As the interpretation of electroencephalography signals, which document brain activity, requires a neurologist’s expertise, this process is often the most expensive component. Therefore, a retrieval system capable of using text to identify relevant signals, eliminating the need for expert analysis, is desirable. Our research proposes a solution to facilitate the creation of indexing systems employing electroencephalography signals for report generation in situations where reports are pending a neurologist review. We introduce a method incorporating a convolutional-neural-network-based encoder from DeepSleepNet, which extracts features from electroencephalography signals, coupled with a transformer which learns the signal’s auto-correlation and the relationship between the signal and the corresponding report. Experimental evaluation using real-world data revealed our approach surpasses baseline methods. These findings suggest potential advancements in medical data retrieval and a decrease in reliance on expert knowledge for electroencephalography signal analysis. As such, our research represents a significant stride towards making electroencephalography data more comprehensible and utilizable in clinical environments.

Classification of electroencephalography using cooperative learning based on participating client balancing

&lt;span lang="EN-US"&gt;Modern technologies are widely used today to diagnose epilepsy, neurological disorders, and brain tumors. Meanwhile, it is not cost-effective in terms of time and money to use a large amount of electroencephalography (EEG) data from different centers and collect them in a central server for processing and analysis. Collecting this data correctly is challenging, and organizations avoid sharing their and client information with others due to data privacy protection. It is difficult to collect these data correctly and it is challenging to transfer them to research centers due to the privacy of the data. In this regard, collaborative learning as an extraordinary approach in this field paves the way for the use of information repositories in research matters without transferring the original data to the centers. This study focuses on the use of a heterogeneous client balancing technique with an interval selection approach and classification of EEG signals with ResNet50 deep architecture. The test results achieved an accuracy of 99.14 compared to similar methods.&lt;/span&gt;

Cognitive control subprocess deficits and compensatory modulation mechanisms in patients with frontal lobe injury revealed by EEG markers: a basic study to guide brain stimulation

BackgroundFrontal lobe injury (FLI) is related to cognitive control impairments, but the influences of FLI on the internal subprocesses of cognitive control remain unclear.AimsWe sought to identify specific biomarkers for...

Association between acute complications in PMM2-CDG patients and haemostasis anomalies: Data from a multicentric study and suggestions for acute management

ObjectivesPatients with PMM2-CDG develop acute events (stroke-like episodes (SLEs), thromboses, haemorrhages, seizures, migraines) associated with both clotting factors (factor XI) and coagulation inhibitors (antithrombin, protein C and protein S) deficiencies. The aim of the study was to correlate acute events to haemostasis and propose practical guidelines. MethodsIn this multicentric retrospective study, we evaluated clinical, radiological, haemostasis and electroencephalography data for PMM2-CDG patients hospitalized for acute events. Cerebral events were classified as thrombosis, haemorrhage, SLE, or “stroke mimic” (SM: normal brain imaging or evoking a migraine). ResultsThirteen patients had a total of 31 acute episodes: 27 cerebral events with 7 SLEs, 4 venous thromboses, 4 haemorrhages (3 associated with thrombosis), 15 SMs at a mean age of 7.7 years; 4 non-cerebral thromboses, one of which included bleeding. A trigger was frequently involved (infection, head trauma). Although sometimes normal at baseline state, factor XI, antithrombin and protein C levels decreased during these episodes. No correlation between haemostasis anomalies and type of acute event was found. DiscussionAcute events in PMM2-CDG are not negligible and are associated with haemostasis anomalies. An emergency protocol is proposed for their prevention and treatment (https://www.filiere-g2m.fr/urgences). For cerebral events, brain Magnetic Resonance Imaging with perfusion weight imaging and diffusion sequences, electroencephalogram and haemostasis protein levels guide the treatment: anticoagulation, antithrombin or fresh frozen plasma supplementation, antiepileptic therapy. Preventing bleeding and thrombosis is required in cases of surgery, prolonged immobilization, hormone replacement therapy. ConclusionAcute events in PMM2-CDG are associated with abnormal haemostasis, requiring practical guidance.

Predicting Dyslexia with Machine Learning: A Comprehensive Review of Feature Selection, Algorithms, and Evaluation Metrics

This literature review explores the use of machine learning-based approaches for the diagnosis and treatment of dyslexia, a learning disorder that affects reading and spelling skills. Various machine learning models, such as artificial neural networks (ANNs), support vector machines (SVMs), and decision trees, have been used to classify individuals as either dyslexic or non-dyslexic based on functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) data. These models have shown promising results for early detection and personalized treatment plans. However, further research is needed to validate these approaches and identify optimal features and models for dyslexia diagnosis and treatment.

Intergenerational transmission of cognitive control capacity among children at risk for depression

A maternal history of major depressive disorder (MDD) is a well-known risk factor for depression in offspring. However, the mechanism through which familial risk is transmitted remains unclear. Cognitive control alterations are common in MDD, and thus, the current study investigated whether altered control capacity is transmitted intergenerationally, and whether it then contributes to the developmental pathways through which depression is passed from mothers to children. We recruited children (N = 65) ages 4–10-years-old, of which 47.7 % (n = 31) reported a maternal history of MDD, and their biological mother (N = 65). Children performed a child-friendly Go/NoGo task while electroencephalography (EEG) data were recorded, and mothers performed a Flanker task. Children exhibited heightened sensitivity to error versus correct responses, which was characterized by an error-related negativity (ERN), error positivity (Pe) as well as prominent delta and frontal midline theta (FMT) oscillations. Interestingly, worse maternal performance on the Flanker task associated with an increased Go/NoGo error rate and a smaller ERN and Pe in children. However, there was no association between maternal or child control indices with child depression symptoms. Our results suggest a familial influence of cognitive control capacity in mother-child dyads, but it remains unclear whether this confers risk for depressive symptoms in children. Further research is necessary to determine whether alterations in cognitive control over time may influence symptom development in at-risk children.

Interrelations and functional roles of key oscillatory activities during daytime sleep in older adults.

Certain neurophysiological characteristics of sleep, in particular slow oscillations (SOs), sleep spindles, and their temporal coupling, have been well characterised and associated with human memory abilities. Delta waves, which are somewhat higher in frequency and lower in amplitude compared to SOs, and their interaction with spindles have only recently been found to play a critical role in memory processing of rodents, through a competitive interaction between SO-spindle and delta-spindle coupling. However, human studies that comprehensively address delta wave interactions with spindles and SOs, as well as their functional role for memory are still lacking. Electroencephalographic data were acquired across three naps of 33 healthy older human participants (17 female) to investigate delta-spindle coupling and the interplay between delta- and SO-related activity. Additionally, we determined intra-individual stability of coupling measures and their potential link to the ability to form novel memories in a verbal memory task. Our results revealed weaker delta-spindle compared to SO-spindle coupling. Contrary to our initial hypothesis, we found no evidence for an opposing dependency between SO- and delta-related activities during non-rapid eye movement sleep. Moreover, the ratio between SO- and delta-nested spindles rather than SO-spindle and delta-spindle coupling measures by themselves predicted the ability to form novel memories best. In conclusion, our results do not confirm previous findings in rodents on competitive interactions between delta activity and SO-spindle coupling in older adults. However, they support the hypothesis that SO, delta wave, and spindle activity should be jointly considered when aiming to link sleep physiology and memory formation.

Enhancing EEG Artifact Removal Efficiency by Introducing Dense Skip Connections to IC-U-Net.

Electroencephalographic (EEG) data is considered contaminated with various types of artifacts. Deep learning has been successfully applied to developing EEG artifact removal techniques to increase the signal-to-noise ratio (SNR) and enhance brain-computer interface performance. Recently, our research team has proposed an end-to-end UNet-based EEG artifact removal technique, IC-U-Net, which can reconstruct signals against various artifacts. However, this model suffers from being prone to overfitting with a limited training dataset size and demanding a high computational cost. To address these issues, this study attempted to leverage the architecture of UNet++ to improve the practicability of IC-U-Net by introducing dense skip connections in the encoder-decoder architecture. Results showed that this proposed model obtained superior SNR to the original model with half the number of parameters. Also, this proposed model achieved comparable convergency using a quarter of the training data size.

A Convolutional Autoencoder-based Explainable Clustering Approach for Resting-State EEG Analysis.

Machine learning methods have frequently been applied to electroencephalography (EEG) data. However, while supervised EEG classification is well-developed, relatively few studies have clustered EEG, which is problematic given the potential for clustering EEG to identify novel subtypes or patterns of dynamics that could improve our understanding of neuropsychiatric disorders. There are established methods for clustering EEG using manually extracted features that reduce the richness of the feature space for clustering, but only a couple studies have sought to use deep learning-based approaches with automated feature learning to cluster EEG. Those studies involve separately training an autoencoder and then performing clustering on the extracted features, and the separation of those steps can lead to poor quality clustering. In this study, we propose an explainable convolutional autoencoder-based approach that combines model training with clustering to yield high quality clusters. We apply the approach within the context of schizophrenia (SZ), identifying 8 EEG states characterized by varying levels of δ activity. We also find that individuals who spend more time outside of the dominant state tend to have increased negative symptom severity. Our approach represents a significant step forward for clustering resting-state EEG data and has the potential to lead to novel findings across a variety of neurological and neuropsychological disorders in future years.

Evaluating methods of oversampling and averaging resting-state electroencephalography data in classifying Parkinson's disease.

Collecting resting-state electroencephalography (RSEEG) data is time-consuming and data sets are therefore often small. Because many machine learning (ML) algorithms work better with ample data, researchers looking to use RSEEG and ML to develop diagnostic models have used oversampling methods that may seem to contradict averaging methods used in conventional electroencephalography (EEG) research to improve the signal-to-noise ratio. Using eyes open (EO) and eyes closed (EC) recordings from 3 different research groups, we investigated the effect of different averaging and oversampling methods on classification metrics when classifying people with Parkinson's disease (PD) and controls. Both EC and EO recordings were used due to differences found between these methods. Our results indicated that grouping 58 electrodes into regions-of-interest (ROI) based on anatomical location is preferable to using single electrodes. Furthermore, although recording EO data led to slightly better classification, the number of data points for each participant was reduced and recordings for three participants entirely lost during pre-processing due to a higher level of artefacts than in the EC data.Clinical relevance- RSEEG is a potential biomarker for the diagnosis and prognostication of PD, but for RSEEG to have clinical relevance, it is necessary to establish which averaging and oversampling of data most reliably segregates the classes for people with PD and controls. We found that using of ROIs and EC data performed the best, as EO data was often contaminated with artefacts.

Children with autism spectrum disorder who demonstrate normal language scores use a bottom-up semantic processing strategy: Evidence from N400 recordings.

The N400 is an electrophysiological component that reflects lexical access and integration of words with mental representations. Thirty-five young children with a range of language capabilities (n =21 neurotypical controls, 10 males, mean age=6.3±0.9 years; n =14 children with autism, 12 males, mean age=6.4±1.1 years) completed an auditory semantic categorization paradigm to evoke the N400. Electroencephalograph (EEG) data were acquired with a 64-channel electrode cap as children listened via ear inserts to binaurally presented single syllable words and decided whether the words were congruent (in) or incongruent (out) with a pre-specified category. EEG data were filtered, epoched, and averaged referenced, and global field power (GFP) was computed. The amplitude of the N400 peak in the GFP was submitted to a multiple linear regression analysis. N400 amplitude was found to predict language scores only for the children with ASD who have language scores in the normal range (r2 =0.72). This finding that N400 amplitude only predicted language scores in children with ASD and normal language scores suggests that these children may rely more on basic semantic processing (as reflected by the N400) and less on anticipating and predicting upcoming words. This suggests preferential utilization of a bottom-up strategy to access higher order language.

Neural correlates of intelligence: ERP Components of temporary storage predict fluid intelligence over and above those of executive functions.

Previous research has revealed that individuals with different levels of intelligence exhibit distinct patterns of event-related potentials (ERPs) related to executive functions and temporary storage. However, there is a lack of studies investigating the relative contributions of these ERPs in predicting individual differences in fluid intelligence. This study aims to examine the extent to which ERPs associated with executive functions and temporary storage can predict individual differences in fluid intelligence. Special attention is given to determining whether electrophysiological activities of temporary storage can predict fluid intelligence after accounting for executive functions, and vice versa. Both executive attention and temporary storage were measured by two experimental tasks, while electroencephalographic data were collected simultaneously. Fluid intelligence was assessed by two established tests. To address previous inconsistencies due to small sample sizes, a relatively large sample of young adults (N = 136) was recruited. The results revealed that participants with lower fluid intelligence displayed larger P3 amplitudes in the executive functions and temporary storage tasks compared to those with higher fluid intelligence. Additionally, the amplitudes of frontal and parietal P3s elicited by both executive functions and temporary storage significantly predicted fluid intelligence. Interestingly, the frontal and parietal P3s associated with temporary storage predicted fluid intelligence beyond the contributions of executive functions, supporting the storage account of individual differences in fluid intelligence. This study provides an original and fresh understanding of how executive functions and temporary storage contribute to fluid intelligence, offering new insights into the neurocognitive mechanisms underlying intelligence.

Social high performers under stress behave more prosocially and detect happy emotions better in a male sample

Psychosocial stress is increasing in society, impacting our lives in all social domains. However, the conditions under which stress facilitates (“tend-and-befriend”) or hinders (“fight-or-flight”) social approach remain elusive. We tested whether previous heterogeneous findings might be resolved by accounting for individual differences in social performance under stress. For that purpose, we introduce the novel Trier Social Stress Test (TSST) social performance index that was aggregated across ratings from two independent observers. Moreover, we apply an innovative setup enabling electroencephalographic (EEG) data to be measured inside an electrically-shielded cabin during stress, namely the TSST-EEG. Relying on a sample of 59 healthy male participants, we collected behavioral (i.e., sharing resources with others) and cognitive (i.e., detecting facial emotional expressions) approach patterns while participants experienced either acute psychosocial stress (n = 31) or no stress (control condition; n = 28) and while EEG was being recorded. During stress exposure, high-performing participants behaved more prosocially, and differentiated better between happy and neutral emotions on both behavioral and neurophysiological levels (revealed by intensity differences in a N170-like response). Overall, our findings demonstrate the added value of both the novel TSST social performance index and the novel TSST-EEG setup. By showing that high social performance during the TSST is associated with behavioral, cognitive, and neurophysiological approach patterns, our study offers valuable insights into adaptive or maladaptive psychobiological mechanisms in coping with psychosocial stress. Future stress research should address the role of social performance differences during stress in social interaction to better understand the behavioral consequences of psychosocial stress in humans.

Blink-related EEG oscillations are neurophysiological indicators of subconcussive head impacts in female soccer players: a preliminary study.

Repetitive subconcussive head impacts can lead to subtle neural changes and functional consequences on brain health. However, the objective assessment of these changes remains limited. Resting state blink-related oscillations (BROs), recently discovered neurological responses following spontaneous blinking, are explored in this study to evaluate changes in BRO responses in subconcussive head impacts. We collected 5-min resting-state electroencephalography (EEG) data from two cohorts of collegiate athletes who were engaged in contact sports (SC) or non-contact sports (HC). Video recordings of all on-field activities were conducted to determine the number of head impacts during games and practices in the SC group. In both groups, we were able to detect a BRO response. Following one season of games and practice, we found a strong association between the number of head impacts sustained by the SC group and increases in delta and beta spectral power post-blink. There was also a significant difference between the two groups in the morphology of BRO responses, including decreased peak-to-peak amplitude of response over left parietal channels and differences in spectral power in delta and alpha frequency range post-blink. Our preliminary results suggest that the BRO response may be a useful biomarker for detecting subtle neural changes resulting from repetitive head impacts. The clinical utility of this biomarker will need to be validated through further research with larger sample sizes, involving both male and female participants, using a longitudinal design.

Secondary motor integration as a final arbiter in sensorimotor decision-making.

Sensorimotor decision-making is believed to involve a process of accumulating sensory evidence over time. While current theories posit a single accumulation process prior to planning an overt motor response, here, we propose an active role of motor processes in decision formation via a secondary leaky motor accumulation stage. The motor leak adapts the "memory" with which this secondary accumulator reintegrates the primary accumulated sensory evidence, thus adjusting the temporal smoothing in the motor evidence and, correspondingly, the lag between the primary and motor accumulators. We compare this framework against different single accumulator variants using formal model comparison, fitting choice, and response times in a task where human observers made categorical decisions about a noisy sequence of images, under different speed-accuracy trade-off instructions. We show that, rather than boundary adjustments (controlling the amount of evidence accumulated for decision commitment), adjustment of the leak in the secondary motor accumulator provides the better description of behavior across conditions. Importantly, we derive neural correlates of these 2 integration processes from electroencephalography data recorded during the same task and show that these neural correlates adhere to the neural response profiles predicted by the model. This framework thus provides a neurobiologically plausible description of sensorimotor decision-making that captures emerging evidence of the active role of motor processes in choice behavior.