Cosine convolutional neural network and its application for seizure detection

Traditional convolutional neural networks (CNNs) often suffer from high memory consumption and redundancy in their kernel representations, leading to overfitting problems and limiting their application in real-time, low-power scenarios such as seizure detection systems. In this work, a novel cosine convolutional neural network (CosCNN), which replaces traditional kernels with the robust cosine kernel modulated by only two learnable factors, is presented, and its effectiveness is validated on the tasks of seizure detection. Meanwhile, based on the cosine lookup table and KL-divergence, an effective post-training quantization algorithm is proposed for CosCNN hardware implementation. With quantization, CosCNN can achieve a nearly 75% reduction in the memory cost with almost no accuracy loss. Moreover, we design a configurable cosine convolution accelerator on Field Programmable Gate Array (FPGA) and deploy the quantized CosCNN on Zedboard, proving the proposed seizure detection system can operate in real-time and low-power scenarios. Extensive experiments and comparisons were conducted using two publicly available epileptic EEG databases, the Bonn database and the CHB-MIT database. The results highlight the performance superiority of the CosCNN over traditional CNNs as well as other seizure detection methods.

Enhancing cross-subject EEG emotion recognition through multi-source manifold metric transfer learning

Transfer learning (TL) has demonstrated its efficacy in addressing the cross-subject domain adaptation challenges in affective brain-computer interfaces (aBCI). However, previous TL methods usually use a stationary distance, such as Euclidean distance, to quantify the distribution dissimilarity between two domains, overlooking the inherent links among similar samples, potentially leading to suboptimal feature mapping. In this study, we introduced a novel algorithm called multi-source manifold metric transfer learning (MSMMTL) to enhance the efficacy of conventional TL. Specifically, we first selected the source domain based on Mahalanobis distance to enhance the quality of the source domains and then used manifold feature mapping approach to map the source and target domains on the Grassmann manifold to mitigate data drift between domains. In this newly established shared space, we optimized the Mahalanobis metric by maximizing the inter-class distances while minimizing the intra-class distances in the target domain. Recognizing that significant distribution discrepancies might persist across different domains even on the manifold, to ensure similar distributions between the source and target domains, we further imposed constraints on both domains under the Mahalanobis metric. This approach aims to reduce distributional disparities and enhance the electroencephalogram (EEG) emotion recognition performance. In cross-subject experiments, the MSMMTL model exhibits average classification accuracies of 88.83 % and 65.04 % for SEED and DEAP, respectively, underscoring the superiority of our proposed MSMMTL over other state-of-the-art methods. MSMMTL can effectively solve the problem of individual differences in EEG-based affective computing.

The use of the eye-fixation-related potential to investigate visual perception in professional domains with high attentional demand: a literature review

INTRODUCTION

Visual attention is a cognitive skill related to visual perception and neural activity, and also moderated by expertise, in time-constrained professional domains (e.g., aviation, driving, sport, surgery). However, the contribution of both perceptual and neural processes on performance has been studied separately in the literature.

DEVELOPMENT

We defend an integration of visual and neural signals to offer a more complete picture of the visual attention displayed by professionals of different skill levels when performing free-viewing tasks. Specifically, we propose to zoom the analysis in data related to the quiet eye and P300 component jointly, as a novel signal processing approach to evaluate professionals' visual attention.

CONCLUSION

This review highlights the advantages of using portable eye trackers and electroencephalogram systems altogether, as a promising technique for a better understanding of early cognitive components related to attentional processes. Altogether, the eye-fixation-related potentials method may provide a better understanding of the cognitive mechanisms employed by the participants in natural settings, revealing what visual information is of interest for participants and distinguishing the neural bases of visual attention between targets and non-targets whenever they perceive a stimulus during free viewing experiments.

A generative adaptive convolutional neural network with attention mechanism for driver fatigue detection with class-imbalanced and insufficient data

Over the past few years, fatigue driving has emerged as one of the main causes of traffic accidents, necessitating the development of driver fatigue detection systems. However, many existing methods involves tedious manual parameter tunings, a process that is both time-consuming and results in task-specific models. On the other hand, most of the researches on fatigue recognition are based on class-balanced and sufficient data, and effectively "mine" meaningful information from class-imbalanced and insufficient data for fatigue recognition is still a challenge. In this paper, we proposed two novel models, the attention-based residual adaptive multiscale fully convolutional network-long short term memory network (ARMFCN-LSTM), and the Generative ARMFCN-LSTM (GARMFCN-LSTM) aiming to address this issue. ARMFCN-LSTM excels at automatically extracting multiscale representations through adaptive multiscale temporal convolutions, while capturing temporal dependency features through LSTM. GARMFCN-LSTM integrates Wasserstein GAN with gradient penalty (WGAN-GP) into ARMFCN-LSTM to improve driver fatigue detection performance by alleviating data scarcity and addressing class imbalances. Experimental results show that ARMFCN-LSTM achieves the highest classification accuracy of 95.84% in driver fatigue detection on the class-balanced EEG dataset (binary classification), and GARMFCN-LSTM attained an improved classification accuracy of 84.70% on the class-imbalanced EOG dataset (triple classification), surpassing the competing methods. Therefore, the proposed models are promising for further implementations in online driver fatigue detection systems.

Differentiating ischemic stroke patients from healthy subjects using a large-scale, retrospective EEG database and machine learning methods

OBJECTIVES

We set out to develop a machine learning model capable of distinguishing patients presenting with ischemic stroke from a healthy cohort of subjects. The model relies on a 3-min resting electroencephalogram (EEG) recording from which features can be computed.

MATERIALS AND METHODS

Using a large-scale, retrospective database of EEG recordings and matching clinical reports, we were able to construct a dataset of 1385 healthy subjects and 374 stroke patients. With subjects often producing more than one recording per session, the final dataset consisted of 2401 EEG recordings (63% healthy, 37% stroke).

RESULTS

Using a rich set of features encompassing both the spectral and temporal domains, our model yielded an AUC of 0.95, with a sensitivity and specificity of 93% and 86%, respectively. Allowing for multiple recordings per subject in the training set boosted sensitivity by 7%, attributable to a more balanced dataset.

CONCLUSIONS

Our work demonstrates strong potential for the use of EEG in conjunction with machine learning methods to distinguish stroke patients from healthy subjects. Our approach provides a solution that is not only timely (3-minutes recording time) but also highly precise and accurate (AUC: 0.95).

Predictors of follow-up care for critically-ill patients with seizures and epileptiform abnormalities on EEG monitoring

OBJECTIVE

Post-hospitalization follow-up visits are crucial for preventing long-term complications. Patients with electrographic epileptiform abnormalities (EA) including seizures and periodic and rhythmic patterns are especially in need of follow-up for long-term seizure risk stratification and medication management. We sought to identify predictors of follow-up.

METHODS

This is a retrospective cohort study of all patients (age ≥ 18 years) admitted to intensive care units that underwent continuous EEG (cEEG) monitoring at a single center between 01/2016-12/2019. Patients with EAs were included. Clinical and demographic variables were recorded. Follow-up status was determined using visit records 6-month post discharge, and visits were stratified as outpatient follow-up, neurology follow-up, and inpatient readmission. Lasso feature selection analysis was performed.

RESULTS

723 patients (53 % female, mean (std) age of 62.3 (16.4) years) were identified from cEEG records with 575 (79 %) surviving to discharge. Of those discharged, 450 (78 %) had outpatient follow-up, 316 (55 %) had a neurology follow-up, and 288 (50 %) were readmitted during the 6-month period. Discharge on antiseizure medications (ASM), younger age, admission to neurosurgery, and proximity to the hospital were predictors of neurology follow-up visits. Discharge on ASMs, along with longer length of stay, younger age, emergency admissions, and higher illness severity were predictors of readmission.

SIGNIFICANCE

ASMs at discharge, demographics (age, address), hospital care teams, and illness severity determine probability of follow-up. Parameters identified in this study may help healthcare systems develop interventions to improve care transitions for critically-ill patients with seizures and other EA.

A hybrid 1D CNN-BiLSTM model for epileptic seizure detection using multichannel EEG feature fusion

Epilepsy - a chronic non communicable condition is characterized by repeated unprovoked seizures, which are transient episodes of abnormal electrical activity in the brain. While Electroencephalography (EEG) is considered as gold standard for diagnosis in current clinical practice, manual inspection of EEG is time taken and biased. This paper presents a novel hybrid 1D CNN-Bi LSTM feature fusion model for automatically detecting seizures. The proposed model leverages spatial features extracted by one dimensional convolutional neural network and temporal features extracted by bi directional long short-term memory network. Ictal and inter ictal data is first acquired from the long multichannel EEG record. The acquired data is segmented and labelled using small fixed windows. Signal features are then extracted from the segments concurrently by the parallel combination of CNN and Bi-LSTM. The spatial and temporal features thus captured are then fused to enhance classification accuracy of model. The approach is validated using benchmark CHB-MIT dataset and 5-fold cross validation resulted in an average accuracy of 95.90%, with precision 94.78%, F1 score 95.95%. Notably model achieved average sensitivity of 97.18% with false positivity rate at 0.05/hr. The significantly lower false positivity and false negativity rates indicate that the proposed model is a promising tool for detecting seizures in epilepsy patients. The employed parallel path network benefits from memory function of Bi-LSTM and strong feature extraction capabilities of CNN. Moreover, eliminating the need for any domain transformation or additional preprocessing steps, model effectively reduces complexity and enhances efficiency, making it suitable for use by clinicians during the epilepsy diagnostic process.

Dynamic decomposition graph convolutional neural network for SSVEP-based brain-computer interface

The SSVEP-based paradigm serves as a prevalent approach in the realm of brain-computer interface (BCI). However, the processing of multi-channel electroencephalogram (EEG) data introduces challenges due to its non-Euclidean characteristic, necessitating methodologies that account for inter-channel topological relations. In this paper, we introduce the Dynamic Decomposition Graph Convolutional Neural Network (DDGCNN) designed for the classification of SSVEP EEG signals. Our approach incorporates layerwise dynamic graphs to address the oversmoothing issue in Graph Convolutional Networks (GCNs), employing a dense connection mechanism to mitigate the gradient vanishing problem. Furthermore, we enhance the traditional linear transformation inherent in GCNs with graph dynamic fusion, thereby elevating feature extraction and adaptive aggregation capabilities. Our experimental results demonstrate the effectiveness of proposed approach in learning and extracting features from EEG topological structure. The results shown that DDGCNN outperforms other state-of-the-art (SOTA) algorithms reported on two datasets (Dataset 1: 54 subjects, 4 targets, 2 sessions; Dataset 2: 35 subjects, 40 targets). Additionally, we showcase the implementation of DDGCNN in the context of synchronized BCI robotic fish control. This work represents a significant advancement in the field of EEG signal processing for SSVEP-based BCIs. Our proposed method processes SSVEP time domain signals directly as an end-to-end system, making it easy to deploy. The code is available at https://github.com/zshubin/DDGCNN.

Impact of intracranial subclinical seizures on seizure outcomes after SLAH in patients with mesial temporal lobe epilepsy

OBJECTIVE

To investigate the association between subclinical seizures detected on intracranial electroencephalographic (i-SCSs)recordings and mesial temporal sclerosis (MTS), as well as their impact on surgical outcomes of stereotactic laser amygdalohippocampotomy (SLAH).

METHODS

A retrospective review was conducted on 27 patients with drug-resistant mesial temporal lobe epilepsy (MTLE) who underwent SLAH. The number of seizures detected on scalp EEG and iEEG was assessed. Patients were followed for a minimum of 3 years after SLAH.

RESULTS

Of the 1715 seizures recorded from mesial temporal regions, 1640 were identified as i-SCSs. Patients with MTS were associated with favorable short- and long-term surgical outcomes. Patients with MTS had a higher number of i-SCSs compared to patients without MTS. The numbers of i-SCSs were higher in patients with Engel I-II outcomes, but no significant statistical difference was found. However, it was observed that patients with MTS who achieved Engel I-II classification had higher numbers of i-SCSs than patients without MTS (P < 0.05).

CONCLUSION

Patients with MTS exhibited favorable short-term and long-term surgical outcome after SLAH. A higher number of i-SCSs was significantly associated with MTS in patients with MTLE. The number of i-SCSs tended to be higher in patients with Engel Ⅰ-Ⅱ surgical outcomes.

SIGNIFICANCE

The association between i-SCSs, MTS, and surgical outcomes in MTLE patients undergoing SLAH has significant implications for understanding the underlying mechanisms and identifying potential therapeutic targets to enhance surgical outcomes.

EEG and EMG-based human-machine interface for navigation of mobility-related assistive wheelchair (MRA-W)

The control of human-machine interfaces (HMIs), such as motorized wheelchairs, has been widely investigated using biopotentials produced by electrochemical processes in the human body. However, many studies in this field sometimes overlook crucial factors like special users' needs, who often have inadequate muscle mass and strength, and paresis needed to operate a wheelchair. This study proposes a novel solution: an economical, universally compatible, and user-centric manual-to-powered wheelchair conversion kit. The powered wheelchair is operated using a hybrid control system integrating electroencephalogram (EEG) and electromyography (EMG), utilizing an LSTM network. It uses a low-cost electroencephalogram (EEG) headset and a wearable electromyography (EMG) electrode armband to solve these constraints. The proposed system comprised three crucial objectives: the development of an EEG-based user attentive detection system, an EMG-based navigation system, and a transform conventional wheelchair into a powered wheelchair. Human test subjects were utilized to evaluate the proposed system, and the study complied with accepted ethical guidelines. We selected four EEG features (p < 0.023) for the attentive detection system and six EMG features (p < 0.037) to detect navigation intentions. User attentive detection was achieved at 83.33 (±0.34) %, while the navigation intention system produced 86.67 (±0.52) % accuracy. The overall system was successful in reaching an accuracy rate of 85.0 (±0.19) % and a weighted average precision of 0.89. After the dataset was trained using an LSTM network, the overall accuracy produced was 97.3 (±0.5) %, higher than the accuracy produced by the Quadratic SVM classifier. By giving older and disabled people a more convenient way to use powered wheelchairs, this research helps to build ergonomic and cost-effective biopotential-based HMIs, enhancing their quality of life.

Simultaneous two-photon imaging and wireless EEG recording in mice

Background

In vivo two-photon imaging is a reliable method with high spatial resolution that allows observation of individual neuron and dendritic activity longitudinally. Neurons in local brain regions can be influenced by global brain states such as levels of arousal and attention that change over relatively short time scales, such as minutes. As such, the scientific rigor of investigating regional neuronal activities could be enhanced by considering the global brain state.

New method

In order to assess the global brain state during in vivo two-photon imaging, CBRAIN (collective brain research platform aided by illuminating neural activity), a wireless EEG collecting and labeling device, was controlled by the same computer of two-photon microscope. In an experiment to explore neuronal responses to isoflurane anesthesia through two-photon imaging, we investigated whether the response of individual cells correlated with concurrent EEG changes induced by anesthesia.

Results

In two-photon imaging, calcium activities of the excitatory neurons in the primary somatosensory cortex disappeared in about 30s after to the initiation of isoflurane anesthesia. The simultaneously recorded EEG showed various transitional activity for about 7 min from the initiation of anesthesia and continued with burst and suppression alternating pattern thereafter. As such, there was a dissociation between excitatory neuron activity of the primary somatosensory cortex and the global brain activity under anesthesia.

Comparison with existing methods

Existing methods to combine two-photon and EEG recording used wired EEG recording. In this study, wireless EEG was used in conjunction with two-photon imaging, facilitated by CBRAIN. More importantly, built-in algorithms of the CBRAIN can automatically detect brain state such as sleep. The codes used for EEG classification are easy to use, with no prior experience required.

Conclusion

Simultaneous recording of wireless EEG and two-photon imaging provides a practical way to capture individual neuronal activities with respect to global brain state in an experimental set-up.

Emotion Recognition with Reduced Channels Using CWT Based EEG Feature Representation and a CNN Classifier

OBJECTIVE

Although emotion recognition has been studied for decades, a more accurate classification method that requires less computing is still needed. At present, in many studies, EEG features are extracted from all channels to recognize emotional states, however, there is a lack of an efficient feature domain that improves classification performance and reduces the number of EEG channels.

APPROACH

In this study, a continuous wavelet transform (CWT)-based feature representation of multi-channel EEG data is proposed for automatic emotion recognition. In the proposed feature, the time-frequency domain information is preserved by using CWT coefficients. For a particular EEG channel, each CWT coefficient is mapped into a strength-to-entropy component ratio to obtain a 2D representation. Finally, a 2D feature matrix, namely CEF2D, is created by concatenating these representations from different channels and fed into a deep convolutional neural network architecture. Based on the CWT domain energy-to-entropy ratio, effective channel and CWT scale selection schemes are also proposed to reduce computational complexity.

MAIN RESULTS

Compared with previous studies, the results of this study show that valence and arousal classification accuracy has improved in both 3-class and 2-class cases. For the 2-class problem, the average accuracies obtained for valence and arousal dimensions are 98.83% and 98.95%, respectively, and for the 3-class, the accuracies are 98.25% and 98.68%, respectively.

SIGNIFICANCE

Our findings show that the entropy-based feature of EEG data in the CWT domain is effective for emotion recognition. Utilizing the proposed feature domain, an effective channel selection method can reduce computational complexity.

A tensor decomposition scheme for EEG-based diagnosis of mild cognitive impairment

Mild Cognitive Impairment (MCI) is the primary stage of acute Alzheimer's disease, and early detection is crucial for the person and those around him. It is difficult to recognize since this mild stage does not have clear clinical signs, and its symptoms are between normal aging and severe dementia. Here, we propose a tensor decomposition-based scheme for automatically diagnosing MCI using Electroencephalogram (EEG) signals. A new projection is proposed, which preserves the spatial information of the electrodes to construct a data tensor. Then, using parallel factor analysis (PARAFAC) tensor decomposition, the features are extracted, and a support vector machine (SVM) is used to discriminate MCI from normal subjects. The proposed scheme was tested on two different datasets. The results showed that the tensor-based method outperformed conventional methods in diagnosing MCI with an average classification accuracy of 93.96% and 78.65% for the first and second datasets, respectively. Therefore, it seems that maintaining the spatial topology of the signals plays a vital role in the processing of EEG signals.

A novel deep-learning model based on τ-shaped convolutional network (τNet) with long short-term memory (LSTM) for physiological fatigue detection from EEG and EOG signals

In recent years, fatigue driving has become the main cause of traffic accidents, leading to increased attention towards fatigue detection systems. However, the pooling and strided convolutional operations in fatigue detection algorithm based on traditional deep learning methods may led to the loss of some useful information. This paper proposed a novel [Formula: see text]-shaped convolutional network ([Formula: see text]) aiming to address this issue. Unlike traditional network structures, [Formula: see text] incorporates the operations of upsampling features and concatenating high- and low-level features, enabling full utilization of useful information. Moreover, considering that the fatigue state is a mental state involving temporal evolution, we proposed the novel long short-term memory (LSTM)-[Formula: see text]-shaped convolutional network (LSTM-[Formula: see text]), a parallel structure composed of LSTM and [Formula: see text] for fatigue detection, where [Formula: see text] extracts time-invariant features with location information, and LSTM extracts long temporal dependencies. We compared LSTM-[Formula: see text] with six competing methods based on two datasets. Results showed that the proposed algorithm achieved higher classification accuracy than the other methods, with 94.25% on EEG data (binary classification) and 82.19% on EOG data (triple classification). Additionally, the proposed algorithm exhibits low computational cost, good training stability, and robustness against insufficient training. Therefore, it is promising for further implementation of fatigue online detection systems.

Prediction of treatment response in major depressive disorder using a hybrid of convolutional recurrent deep neural networks and effective connectivity based on EEG signal

In this study, we have developed a novel method based on deep learning and brain effective connectivity to classify responders and non-responders to selective serotonin reuptake inhibitors (SSRIs) antidepressants in major depressive disorder (MDD) patients prior to the treatment using EEG signal. The effective connectivity of 30 MDD patients was determined by analyzing their pretreatment EEG signals, which were then concatenated into delta, theta, alpha, and beta bands and transformed into images. Using these images, we then fine tuned a hybrid Convolutional Neural Network that is enhanced with bidirectional Long Short-Term Memory cells based on transfer learning. The Inception-v3, ResNet18, DenseNet121, and EfficientNet-B0 models are implemented as base models. Finally, the models are followed by BiLSTM and dense layers in order to classify responders and non-responders to SSRI treatment. Results showed that the EfficiencyNet-B0 has the highest accuracy of 98.33, followed by DensNet121, ResNet18 and Inception-v3. Therefore, a new method was proposed in this study that uses deep learning models to extract both spatial and temporal features automatically, which will improve classification results. The proposed method provides accurate identification of MDD patients who are responding, thereby reducing the cost of medical facilities and patient care.

Electro-clinical features and long-term outcomes in guanidinoacetate methyltransferase (GAMT) deficiency

OBJECTIVE

To evaluate clinical characteristics and long-term outcomes in patients with guanidinoacetate methyltransferase (GAMT) deficiency with a special emphasis on seizures and electroencephalography (EEG) findings.

METHODS

We retrospectively analyzed the clinical and molecular characteristics, seizure types, EEG findings, neuroimaging features, clinical severity scores, and treatment outcomes in six patients diagnosed with GAMT deficiency.

RESULTS

Median age at presentation and diagnosis were 11.5 months (8-12 months) and 63 months (18 months -11 years), respectively. Median duration of follow-up was 14 years. Global developmental delay (6/6) and seizures (5/6) were the most common symptoms. Four patients presented with febrile seizures. The age at seizure-onset ranged between 8 months and 4 years. Most common seizure types were generalized tonic seizures (n = 4) and motor seizures resulting in drop attacks (n = 3). Slow background activity (n = 5) and generalized irregular sharp and slow waves (n = 3) were the most common EEG findings. Burst-suppression and electrical status epilepticus during slow-wave sleep (ESES) pattern was present in one patient. Three of six patients had drug-resistant epilepsy. Post-treatment clinical severity scores showed improvement regarding movement disorders and epilepsy. All patients were seizure-free in the follow-up.

CONCLUSIONS

Epilepsy is one of the main symptoms in GAMT deficiency with various seizure types and non-specific EEG findings. Early diagnosis and initiation of treatment are crucial for better seizure and cognitive outcomes. This long-term follow up study highlights to include cerebral creatine deficiency syndromes in the differential diagnosis of patients with global developmental delay and epilepsy and describes the course under treatment.

BrainGridNet: A two-branch depthwise CNN for decoding EEG-based multi-class motor imagery

Brain-computer interfaces (BCIs) based on motor imagery (MI) enable the disabled to interact with the world through brain signals. To meet demands of real-time, stable, and diverse interactions, it is crucial to develop lightweight networks that can accurately and reliably decode multi-class MI tasks. In this paper, we introduce BrainGridNet, a convolutional neural network (CNN) framework that integrates two intersecting depthwise CNN branches with 3D electroencephalography (EEG) data to decode a five-class MI task. The BrainGridNet attains competitive results in both the time and frequency domains, with superior performance in the frequency domain. As a result, an accuracy of 80.26 percent and a kappa value of 0.753 are achieved by BrainGridNet, surpassing the state-of-the-art (SOTA) model. Additionally, BrainGridNet shows optimal computational efficiency, excels in decoding the most challenging subject, and maintains robust accuracy despite the random loss of 16 electrode signals. Finally, the visualizations demonstrate that BrainGridNet learns discriminative features and identifies critical brain regions and frequency bands corresponding to each MI class. The convergence of BrainGridNet's strong feature extraction capability, high decoding accuracy, steady decoding efficacy, and low computational costs renders it an appealing choice for facilitating the development of BCIs.

Exploring recurrence quantification analysis and fractal dimension algorithms for diagnosis of encephalopathy

Electroencephalography (EEG) is a crucial non-invasive medical tool for diagnosing neurological disorder called encephalopathy. There is a requirement for powerful signal processing algorithms as EEG patterns in encephalopathies are not specific to a particular etiology. As visual examination and linear methods of EEG analysis are not sufficient to get the subtle information regarding various neuro pathologies, non-linear analysis methods can be employed for exploring the dynamic, complex and chaotic nature of EEG signals. This work aims identifying and differentiating the patterns specific to cerebral dysfunctions associated with Encephalopathy using Recurrence Quantification Analysis and Fractal Dimension algorithms. This study analysed six RQA features, namely, recurrence rate, determinism, laminarity, diagonal length, diagonal entropy and trapping time and comparing them with fractal dimensions, namely, Higuchi's and Katz's fractal dimension. Fractal dimensions were found to be lower for encephalopathy cases showing decreased complexity when compared to that of normal healthy subjects. On the other hand, RQA features were found to be higher for encephalopathy cases indicating higher recurrence and more periodic patterns in EEGs of encephalopathy compared to that of normal healthy controls. The feature reduction was then performed using Principal Component Analysis and fed to three promising classifiers: SVM, Random Forest and Multi-layer Perceptron. The resultant system provides a practically realizable pipeline for the diagnosis of encephalopathy.

A Personalized Semi-Automatic Sleep Spindle Detection (PSASD) Framework

BACKGROUND

Sleep spindles are distinct electroencephalogram (EEG) patterns of brain activity that have been posited to play a critical role in development, learning, and neurological disorders. Manual scoring for sleep spindles is labor-intensive and tedious but could supplement automated algorithms to resolve challenges posed with either approaches alone.

NEW METHODS

A Personalized Semi-Automatic Sleep Spindle Detection (PSASD) framework was developed to combine the strength of automated detection algorithms and visual expertise of human scorers. The underlying model in the PSASD framework assumes a generative model for EEG sleep spindles as oscillatory components, optimized to EEG amplitude, with remaining signals distributed into transient and low-frequency components.

RESULTS

A single graphical user interface (GUI) allows both manual scoring of sleep spindles (model training data) and verification of automatically detected spindles. A grid search approach allows optimization of parameters to balance tradeoffs between precision and recall measures.

COMPARISON WITH EXISTING METHODS

PSASD outperformed DETOKS in F1-score by 19% and 4% on the DREAMS and P-DROWS-E datasets, respectively. It also outperformed YASA in F1-score by 25% in the P-DROWS-E dataset. Further benchmarking analysis showed that PSASD outperformed four additional widely used sleep spindle detectors in F1-score in the P-DROWS-E dataset. Titration analysis revealed that four 30-second epochs are sufficient to fine-tune the model parameters of PSASD. Associations of frequency, duration, and amplitude of detected sleep spindles matched those previously reported with automated approaches.

CONCLUSIONS

Overall, PSASD improves detection of sleep spindles in EEG data acquired from both younger healthy and older adult patient populations.

Evaluating the effectiveness of artifact correction and rejection in event-related potential research

Eyeblinks and other large artifacts can create two major problems in event-related potential (ERP) research, namely confounds and increased noise. Here, we developed a method for assessing the effectiveness of artifact correction and rejection methods at minimizing these two problems. We then used this method to assess a common artifact minimization approach, in which independent component analysis (ICA) is used to correct ocular artifacts, and artifact rejection is used to reject trials with extreme values resulting from other sources (e.g., movement artifacts). This approach was applied to data from five common ERP components (P3b, N400, N170, mismatch negativity, and error-related negativity). Four common scoring methods (mean amplitude, peak amplitude, peak latency, and 50% area latency) were examined for each component. We found that eyeblinks differed systematically across experimental conditions for several of the components. We also found that artifact correction was reasonably effective at minimizing these confounds, although it did not usually eliminate them completely. In addition, we found that the rejection of trials with extreme voltage values was effective at reducing noise, with the benefits of eliminating these trials outweighing the reduced number of trials available for averaging. For researchers who are analyzing similar ERP components and participant populations, this combination of artifact correction and rejection approaches should minimize artifact-related confounds and lead to improved data quality. Researchers who are analyzing other components or participant populations can use the method developed in this study to determine which artifact minimization approaches are effective in their data.

Physiological measurements for driving drowsiness: A comparative study of multi-modality feature fusion and selection

A large number of traffic accidents were caused by drowsiness while driving. In-vehicle alert system based on physiological signals was one of the most promising solutions to monitor driving fatigue. However, different physiological modalities can be used, and many relative studies compared different modalities without considering the implementation feasibility of portable or wearable devices. Moreover, evaluations of each modality in previous studies were based on inconsistent choices of fatigue label and signal features, making it hard to compare the results of different studies. Therefore, the modality comparison and fusion for continuous drowsiness estimation while driving was still unclear. This work sought to comprehensively compare widely-used physiological modalities, including forehead electroencephalogram (EEG), electrooculogram (EOG), R-R intervals (RRI) and breath, in a hardware setting feasible for portable or wearable devices to monitor driving fatigue. Moreover, a more general conclusion on modality comparison and fusion was reached based on the regression of features or their combinations and the awake-to-drowsy transition. Finally, the feature subset of fused modalities was produced by feature selection method, to select the optimal feature combination and reduce computation consumption. Considering practical feasibility, the most effective combination with the highest correlation coefficient was using forehead EEG or EOG, along with RRI and RRI-derived breath. If more comfort and convenience was required, the combination of RRI and RRI-derived breath was also promising.

Neural effects of TMS trains on the human prefrontal cortex

How does a train of TMS pulses modify neural activity in humans? Despite adoption of repetitive TMS (rTMS) for the treatment of neuropsychiatric disorders, we still do not understand how rTMS changes the human brain. This limited understanding stems in part from a lack of methods for noninvasively measuring the neural effects of a single TMS train - a fundamental building block of treatment - as well as the cumulative effects of consecutive TMS trains. Gaining this understanding would provide foundational knowledge to guide the next generation of treatments. Here, to overcome this limitation, we developed methods to noninvasively measure causal and acute changes in cortical excitability and evaluated this neural response to single and sequential TMS trains. In 16 healthy adults, standard 10 Hz trains were applied to the dorsolateral prefrontal cortex (dlPFC) in a randomized, sham-controlled, event-related design and changes were assessed based on the TMS-evoked potential (TEP), a measure of cortical excitability. We hypothesized that single TMS trains would induce changes in the local TEP amplitude and that those changes would accumulate across sequential trains, but primary analyses did not indicate evidence in support of either of these hypotheses. Exploratory analyses demonstrated non-local neural changes in sensor and source space and local neural changes in phase and source space. Together these results suggest that single and sequential TMS trains may not be sufficient to modulate local cortical excitability indexed by typical TEP amplitude metrics but may cause neural changes that can be detected outside the stimulation area or using phase or source space metrics. This work should be contextualized as methods development for the monitoring of transient noninvasive neural changes during rTMS and contributes to a growing understanding of the neural effects of rTMS.

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.

Automatic identification of schizophrenia employing EEG records analyzed with deep learning algorithms

Electroencephalography is a method of detecting and analyzing electrical activity in the brain. This electrical activity can be recorded and processed to aid in the clinical diagnosis of mental disorders. In this study, a novel system for classifying schizophrenia patients from EEG recordings is presented. The developed algorithm decomposes the EEG signals into a system of radial basis functions using the method of fuzzy means. This decomposition helps to obtain the information from the various electrodes of the EEG and allows separating between healthy controls and patients with schizophrenia. The proposed method has been compared with classical machine learning algorithms, such as, K-Nearest Neighbor, Adaboost, Support Vector Machine, and Bayesian Linear Discriminant Analysis. The results show that the proposed method obtains the highest values in terms of balanced accuracy, recall, precision and F1 score, close to 93 % in all cases. The model developed in this study can be implemented in brain activity analysis systems that help in the prediction of patients with schizophrenia.

Different patterns of foreground and background processing contribute to texture segregation in humans: an electrophysiological study

Background

Figure-ground segregation is a necessary process for accurate visual recognition. Previous neurophysiological and human brain imaging studies have suggested that foreground-background segregation relies on both enhanced foreground representation and suppressed background representation. However, in humans, it is not known when and how foreground and background processing play a role in texture segregation.

Methods

To answer this question, it is crucial to extract and dissociate the neural signals elicited by the foreground and background of a figure texture with high temporal resolution. Here, we combined an electroencephalogram (EEG) recording and a temporal response function (TRF) approach to specifically track the neural responses to the foreground and background of a figure texture from the overall EEG recordings in the luminance-tracking TRF. A uniform texture was included as a neutral condition. The texture segregation visual evoked potential (tsVEP) was calculated by subtracting the uniform TRF from the foreground and background TRFs, respectively, to index the specific segregation activity.

Results

We found that the foreground and background of a figure texture were processed differently during texture segregation. In the posterior region of the brain, we found a negative component for the foreground tsVEP in the early stage of foreground-background segregation, and two negative components for the background tsVEP in the early and late stages. In the anterior region, we found a positive component for the foreground tsVEP in the late stage, and two positive components for the background tsVEP in the early and late stages of texture processing.

Discussion

In this study we investigated the temporal profile of foreground and background processing during texture segregation in human participants at a high time resolution. The results demonstrated that the foreground and background jointly contribute to figure-ground segregation in both the early and late phases of texture processing. Our findings provide novel evidence for the neural correlates of foreground-background modulation during figure-ground segregation in humans.

An explainable deep-learning model to stage sleep states in children and propose novel EEG-related patterns in sleep apnea

Automatic deep-learning models used for sleep scoring in children with obstructive sleep apnea (OSA) are perceived as black boxes, limiting their implementation in clinical settings. Accordingly, we aimed to develop an accurate and interpretable deep-learning model for sleep staging in children using single-channel electroencephalogram (EEG) recordings. We used EEG signals from the Childhood Adenotonsillectomy Trial (CHAT) dataset (n = 1637) and a clinical sleep database (n = 980). Three distinct deep-learning architectures were explored to automatically classify sleep stages from a single-channel EEG data. Gradient-weighted Class Activation Mapping (Grad-CAM), an explainable artificial intelligence (XAI) algorithm, was then applied to provide an interpretation of the singular EEG patterns contributing to each predicted sleep stage. Among the tested architectures, a standard convolutional neural network (CNN) demonstrated the highest performance for automated sleep stage detection in the CHAT test set (accuracy = 86.9% and five-class kappa = 0.827). Furthermore, the CNN-based estimation of total sleep time exhibited strong agreement in the clinical dataset (intra-class correlation coefficient = 0.772). Our XAI approach using Grad-CAM effectively highlighted the EEG features associated with each sleep stage, emphasizing their influence on the CNN's decision-making process in both datasets. Grad-CAM heatmaps also allowed to identify and analyze epochs within a recording with a highly likelihood to be misclassified, revealing mixed features from different sleep stages within these epochs. Finally, Grad-CAM heatmaps unveiled novel features contributing to sleep scoring using a single EEG channel. Consequently, integrating an explainable CNN-based deep-learning model in the clinical environment could enable automatic sleep staging in pediatric sleep apnea tests.

Effect of Lacosamide on Interictal Epileptiform Discharges in Pediatric Patients With Newly Diagnosed Focal Epilepsy

BACKGROUND

The purpose of this study was to determine the efficacy of lacosamide (LCM) on interictal epileptiform discharges (IEDs) and evaluate the relationships between IEDs and seizure outcome in pediatric patients with focal epilepsy.

METHODS

Patient inclusion criteria included (1) newly diagnosed focal epilepsy with unknown etiology; and (2) electroencephalogram recorded twice (before and after starting LCM) under the same conditions. The difference between the highest number of IEDs over five successive minutes (IEDs/5 min) and the location of IEDs was determined. Seizure outcome was evaluated one year after achieving the maintenance dose of LCM. Responders were identified as showing a ≥50% reduction in the pre-LCM seizure frequency.

RESULTS

Of 22 patients, 10 showed an increase in IEDs/5 min after starting LCM. The median IEDs/5 min before and after starting LCM was not significantly different, at 1.5 (interquartile range: 0, 31.75) and 10.5 (0, 80.5), respectively. No relationship was identified between the difference in IEDs/5 min and seizure outcome. Patients with multiple regional or diffuse IEDs had significantly poorer seizure outcome compared with patients without those IEDs (P = 0.036 and P = 0.039, respectively). Of 10 patients with single regional IEDs, a tendency of IEDs to disappear was observed between patients with frontal and non-frontal IEDs.

CONCLUSION

The effects of LCM on the number of IEDs may be unrelated to seizure outcome. LCM may be ineffective at improving seizure outcomes in patients with multiple regional or diffuse IEDs.

An Overview on Electrophysiological and Neuroimaging Findings in Dyslexia

Objective: Dyslexia is a prevalent neurodevelopmental condition that is characterized by inaccurate and slow word recognition. This article reviews neural correlates of dyslexia from both electrophysiological and neuroimaging studies. Method : In this brief review, we provide electrophysiological and neuroimaging evidence from electroencephalogram (EEG) and magnetic resonance imaging (MRI) studies in dyslexia to understand functional and structural brain changes in this condition. Results: In both electrophysiological and neuroimaging studies, the most frequently reported functional impairments in dyslexia include aberrant activation of the left hemisphere occipito-temporal cortex (OTC), temporo-parietal cortex (TPC), inferior frontal gyrus (IFG), and cerebellar areas. EEG studies have mostly highlighted the important role of lower frequency bands in dyslexia, especially theta waves. Furthermore, neuroimaging studies have suggested that dyslexia is related to functional and structural impairments in the left hemisphere regions associated with reading and language, including reduced grey matter volume in the left TPC, decreased white matter connectivity between reading networks, and hypo-activation of the left OTC and TPC. In addition, neural evidence from pre-reading children and infants at risk for dyslexia show that there are abnormalities in the dyslexic brain before learning to read begins. Conclusion: Advances in comprehending the neural correlates of dyslexia could bring closer translation from basic to clinical neuroscience and effective rehabilitation for individuals who struggle to read. However, neuroscience still has great potential for clinical translation that requires further research.

Neurobiological effects of exercise intervention for premenstrual syndrome

Background

Up to 75%-90% of women have varying degrees of premenstrual syndrome (PMS). Exercises are recognized to be beneficial to regulate the negative emotions associated with PMS; however, the effects of exercise on sadness inhibition have not yet been investigated from the neurobiological perspective.

Purpose

This study examined the effects of a single exercise intervention on the neural mechanisms mediating sadness response inhibition at the cortical level using multichannel event-related potential (ERP) recording in women with PMS.

Methods

Participants performed Go/No-go trials while viewing of sad or neutral images before and after exercise intervention, and changes in the No-go-evoked N200 (N2) ERP component were measured by electroencephalography (EEG) at multiple cortical sites. The associations of PMS Inventory scores with N2 amplitude and latency changes were then examined using Pearson's correlation analysis.

Results

There were no significant differences in N2 latency and response error rate following exercise compared to baseline. However, women with higher PMS Inventory scores (greater symptom severity) demonstrated significantly lengthen N2 latency at the Fz electrode sites during correct sad face No-go trials after exercise (p < 0.05), which was not the case in the pre-exercise baseline. We detected no significant relationship between the PMS score and N2 amplitude, either pre- or post-exercise.

Conclusion

Women with higher PMS severity exhibited longer sad N2 latencies as well as slow down the speed of reaction to negative stimuli by exercise, suggesting that the prefrontal emotion regulation network is involved in PMS symptoms and is sensitive to the beneficial effects of exercise.

Self-attention-based convolutional neural network and time-frequency common spatial pattern for enhanced motor imagery classification

BACKGROUND

Motor imagery (MI) based brain-computer interfaces (BCIs) have promising potentials in the field of neuro-rehabilitation. However, due to individual variations in active brain regions during MI tasks, the challenge of decoding MI EEG signals necessitates improved classification performance for practical application.

NEW METHOD

This study proposes a self-attention-based Convolutional Neural Network (CNN) in conjunction with a time-frequency common spatial pattern (TFCSP) for enhanced MI classification. Due to the limited availability of training data, a data augmentation strategy is employed to expand the scale of MI EEG datasets. The self-attention-based CNN is trained to automatically extract the temporal and spatial information from EEG signals, allowing the self-attention module to select active channels by calculating EEG channel weights. TFCSP is further implemented to extract multiscale time-frequency-space features from EEG data. Finally, the EEG features derived from TFCSP are concatenated with those from the self-attention-based CNN for MI classification.

RESULTS

The proposed method is evaluated on two publicly accessible datasets, BCI Competition IV IIa and BCI Competition III IIIa, yielding mean accuracies of 79.28 % and 86.39 %, respectively.

CONCLUSIONS

Compared with state-of-the-art methods, our approach achieves superior classification results in accuracy. Self-attention-based CNN combining with TFCSP can make full use of the time-frequency-space information of EEG, and enhance the classification performance.

EEG-based emotion recognition using dual tree complex wavelet transform and random subspace ensemble classifier

Emotions are strongly admitted as a main source to establish meaningful interactions between humans and computers. Thanks to the advancements in electroencephalography (EEG), especially in the usage of portable and cheap wearable EEG devices, the demand for identifying emotions has extremely increased. However, the overall scientific knowledge and works concerning EEG-based emotion recognition is still limited. To cover this issue, we introduce an EEG-based emotion recognition framework in this study. The proposed framework involves the following stages: preprocessing, feature extraction, feature selection and classification. For the preprocessing stage, multi scale principle component analysis and sysmlets-4 filter are used. A version of discrete wavelet transform (DWT), namely dual tree complex wavelet transform (DTCWT) is utilized for the feature extraction stage. To reduce the feature dimension size, a variety of statistical criteria are employed. For the final stage, we adopt ensemble classifiers due to their promising performance in classification problems. The proposed framework achieves nearly 96.8% accuracy by using random subspace ensemble classifier. It can therefore be resulted that the proposed EEG-based framework performs well in terms of identifying emotions.

Discrete cortical control during quiet stance revealed by desynchronization and rebound of beta oscillations

Postural sway during quiet stance often exhibits a repetition of micro-fall and the subsequent micro-recovery. The classical view -that the quiet bipedal stance is stabilized by the ankle joint stiffness- has been challenged by paradoxical non-spring-like behaviors of calf muscles: gastrocnemius muscles are shortened and then lengthened, respectively, during the micro-fall and the micro-recovery. Here, we examined EEG based brain activity during quiet stance, and identified desynchronization and synchronization of beta oscillations that were associated, respectively, with the micro-fall and the micro-recovery. Based on a widely accepted scenario for beta-band desynchronization during movement and post-movement rebound in the control of discrete voluntary movement, our results reveal that the beta rebound can be considered as a manifestation of stop command to punctuate the motor control for every fall-recovery cycle. Namely, cortical interventions to the automatic postural control are discrete, rather than continuous modulations. The finding is highly compatible with the intermittent control model, rather than the stiffness control model.

Spontaneous brain microstates correlate with impaired inhibitory control in internet addiction disorder

The prevalence of the Internet addiction disorder (IAD) has been on the rise, making it increasingly imperative to explore the neurophysiological markers of it. Using the whole-brain imaging approach of EEG microstate analysis, which treats multichannel EEG recordings as a series of quasi-steady states, similar as the resting-state networks found by fMRI, the present study aimed to investigate the specificity of the IAD in class C of the four canonical microstates. The existing EEG data of 40 participants (N = 20 for each group) was used, and correlation between the time parameters of microstate C and the performance of the Go/NoGo task was analyzed. Results suggested that the duration and coverage of class C were significantly reduced in the IAD group as compared to the healthy control (HC) group. Furthermore, the duration of class C had a significant inverse correlation with Go RTs in the IAD group. These results implied that class C might serve as a neurophysiological marker of IAD, helping to understand the underlying neural mechanism of inhibitory control in IAD.

Cueing effect of attention among nurses with different anxiety levels: an EEG study

The attention to cueing among nurses with anxiety affects their nursing quality seriously. Nevertheless, the neural mechanism of attention under anxiety among nurses has not been revealed. In this study, we utilized the event-related potential (ERP) and functional brain networks to investigate the neural mechanism of the cueing attention differences between anxiety and non-anxiety nurse groups (AG-20 nurses; NAG-20 nurses) in the spatial cueing task. The results revealed that in the invalid cues (144 trials), longer reaction times, larger P2 amplitudes, and more linkages between the right frontal and parietal areas were found in AG compared to NAG. In the valid cues (288 trials), there were no significant behavioral and neural differences between the two groups. The AG in the invalid cues showed slower response times, larger P2 and N5 amplitudes, and denser linkages originating from the occipital cortex than those in the valid cues. The convolutional neural network was trained for discriminating between the anxiety nurses and the normal ones, with the average accuracy being 0.76. The findings provided a potential physiological biomarker to predict the anxiety group who need to give more psychological attention. Nurse leaders maybe get more information for offering solutions to retain mental health among nurses.

Enhancement of dynamic visual acuity using transcranial alternating current stimulation with gamma burst entrained on alpha wave troughs

BACKGROUND

Cross-frequency phase-amplitude coupling (PAC) of cortical oscillations is observed within and across cortical regions during higher-order cognitive processes. Particularly, the PAC of alpha and gamma waves in the occipital cortex is closely associated with visual perception. In theory, gamma oscillation is a neuronal representation of visual stimuli, which drives the duty cycle of visual perception together with alpha oscillation. Therefore, it is believed that the timing of entrainment in alpha-gamma PAC may play a critical role in the performance of visual perception. We hypothesized that transcranial alternating current stimulation (tACS) with gamma waves entrained at the troughs of alpha waves would enhance the dynamic visual acuity (DVA).

METHOD

We attempted to modulate the performance of DVA by using tACS. The waveforms of the tACS were tailored to target PAC over the occipital cortex. The waveforms contained gamma (80 Hz) waves oscillating at either the peaks or troughs of alpha (10 Hz) waves. Participants performed computerized DVA task before, immediately after, and 10 min after each stimulation sessions. EEG and EOG were recorded during the DVA task to assess inter-trial phase coherence (ITPC), the alpha-gamma PAC at occipital site and the eye movements.

RESULTS

tACS with gamma waves entrained at alpha troughs effectively enhanced DVA, while the tACS with gamma waves entrained at alpha peaks did not affect DVA performance. Importantly, analyses of EEG and EOG showed that the enhancement of DVA performance originated solely from the neuromodulatory effects, and was not related to the modulation of saccadic eye movements. Consequently, DVA, one of the higher-order cognitive abilities, was successfully modulated using tACS with a tailored waveform.

CONCLUSIONS

Our experimental results demonstrated that DVA performances were enhanced when tACS with gamma bursts entrained on alpha wave troughs were applied over the occipital cortex. Our findings suggest that using tACS with tailored waveforms, modulation of complex neuronal features could effectively enhance higher-order cognitive abilities such as DVA, which has never been modulated with conventional noninvasive brain stimulation methods.

A novel noninvasive brain-computer interface by imagining isometric force levels

Physiological circuits differ across increasing isometric force levels during unilateral contraction. Therefore, we first explored the possibility of predicting the force level based on electroencephalogram (EEG) activity recorded during a single trial of unilateral 5% or 40% of maximal isometric voluntary contraction (MVC) in right-hand grip imagination. Nine healthy subjects were involved in this study. The subjects were required to randomly perform 20 trials for each force level while imagining a right-hand grip. We proposed the use of common spatial patterns (CSPs) and coherence between EEG signals as features in a support vector machine for force level prediction. The results showed that the force levels could be predicted through single-trial EEGs while imagining the grip (mean accuracy = 81.4 ± 13.29%). Additionally, we tested the possibility of online control of a ball game using the above paradigm through unilateral grip imagination at different force levels (i.e., 5% of MVC imagination and 40% of MVC imagination for right-hand movement control). Subjects played the ball games effectively by controlling direction with our novel BCI system (n = 9, mean accuracy = 76.67 ± 9.35%). Data analysis validated the use of our BCI system in the online control of a ball game. This information may provide additional commands for the control of robots by users through combinations with other traditional brain-computer interfaces, e.g., different limb imaginations.

EEG-based emergency braking intention detection during simulated driving

BACKGROUND

Current research related to electroencephalogram (EEG)-based driver's emergency braking intention detection focuses on recognizing emergency braking from normal driving, with little attention to differentiating emergency braking from normal braking. Moreover, the classification algorithms used are mainly traditional machine learning methods, and the inputs to the algorithms are manually extracted features.

METHODS

To this end, a novel EEG-based driver's emergency braking intention detection strategy is proposed in this paper. The experiment was conducted on a simulated driving platform with three different scenarios: normal driving, normal braking and emergency braking. We compared and analyzed the EEG feature maps of the two braking modes, and explored the use of traditional methods, Riemannian geometry-based methods, and deep learning-based methods to predict the emergency braking intention, all using the raw EEG signals rather than manually extracted features as input.

RESULTS

We recruited 10 subjects for the experiment and used the area under the receiver operating characteristic curve (AUC) and F1 score as evaluation metrics. The results showed that both the Riemannian geometry-based method and the deep learning-based method outperform the traditional method. At 200 ms before the start of real braking, the AUC and F1 score of the deep learning-based EEGNet algorithm were 0.94 and 0.65 for emergency braking vs. normal driving, and 0.91 and 0.85 for emergency braking vs. normal braking, respectively. The EEG feature maps also showed a significant difference between emergency braking and normal braking. Overall, based on EEG signals, it was feasible to detect emergency braking from normal driving and normal braking.

CONCLUSIONS

The study provides a user-centered framework for human-vehicle co-driving. If the driver's intention to brake in an emergency can be accurately identified, the vehicle's automatic braking system can be activated hundreds of milliseconds earlier than the driver's real braking action, potentially avoiding some serious collisions.

Brain activity differences between susceptible and non-susceptible populations under visually induced motion sickness based on sensor-space and source-space analyses

The neural mechanisms underlying visually induced motion sickness (VIMS) in different susceptible populations are unclear, as it is not clear how brain activity changes in different susceptible populations during the vection section (VS). This study aimed to analyze the brain activity changes in different susceptible populations during VS. Twenty subjects were included in this study and divided into the VIMS-susceptible group (VIMSSG) and VIMS-resistant group (VIMSRG) based on a motion sickness questionnaire. 64-channel electroencephalogram (EEG) data from these subjects during VS were collected. The brain activities during VS for VIMSSG and VIMSRG were analyzed with time-frequency based sensor-space analysis and EEG source imaging based source-space analysis. Under VS, delta and theta energies were significantly increased in VIMSSG and VIMSRG, while alpha and beta energies were only significantly increased in VIMSRG. Also, the superior and middle temporal were activated in VIMSSG and VIMSRG, while lateral occipital, supramarginal gyrus, and precentral gyrus were activated only in VIMSSG. The spatiotemporal differences in brain activity observed between VIMSSG and VIMSRG may be attributed to the different susceptibility of participants in each group and the different severity of MS symptoms experienced. Long-term vestibular training can effectively improve the ability of anti-VIMS. The knowledge gained from this study helps advance understanding of the neural mechanism of VIMS in different susceptible populations.

Distance to Criticality Undergoes Critical Transition Before Epileptic Seizure Attacks

Epilepsy is a common neurological disorder characterized by recurring seizures, but its underlying mechanisms remain poorly understood. Despite extensive research, there are still gaps in our knowledge about the relationship between brain dynamics and seizures. In this study, our aim is to address these gaps by proposing a novel approach to assess the role of brain network dynamics in the onset of seizures. Specifically, we investigate the relationship between brain dynamics and seizures by tracking the distance to criticality. Our hypothesis is that this distance plays a crucial role in brain state changes and that seizures may be related to critical transitions of this distance. To test this hypothesis, we develop a method to measure the evolution of the brain network's distance to the critical dynamic systems (i.e., the distance to the tipping point, DTP) using dynamic network biomarker theory and random matrix theory. The results show that the DTP of the brain decreases significantly immediately after onset of an epileptic seizure, suggesting that the brain loses its well-defined quasi-critical state during seizures. We refer to this phenomenon as the "criticality of the criticality" (COC). Furthermore, we observe that DTP exhibits a shape transition before and after the onset of the seizures. This phenomenon suggests the possibility of early warning signal (EWS) identification in the dynamic sequence of DTP, which could be utilized for seizure prediction. Our results show that the Hurst exponent, skewness, kurtosis, autocorrelation, and variance of the DTP sequence are potential EWS features. This study advances our understanding of the relationship between brain dynamics and seizures and highlights the potential for using criticality-based measures to predict and prevent seizures.

Quantitative and Qualitative Representation of Introductory and Advanced EEG Concepts: An Exploration of Different EEG Setups

Electroencephalograms (EEGs) are the gold standard test used in the medical field to diagnose epilepsy and aid in the diagnosis of many other neurological and mental disorders. Growing in popularity in terms of nonmedical applications, the EEG is also used in research, neurofeedback, and brain-computer interface, making it increasingly relevant to student learning. Recent innovations have made EEG setups more accessible and affordable, thus allowing their integration into neuroscience educational settings. Introducing students to EEGs, however, can be daunting due to intricate setup protocols, individual variation, and potentially expensive equipment. This paper aims to provide guidance for introducing students and educators to fundamental beginning and advanced level EEG concepts. Specifically, this paper tested the potential of three different setups, with varying channel number and wired or wireless connectivity, for introducing students to qualitative and quantitative exploration of alpha enhancement when eyes are closed, and observation of the alpha/beta anterior to posterior gradient. The setups were compared to determine their relative advantages and their robustness in detecting these well-established parameters. The basic 1- or 2-channel setups are sufficient for observing alpha and beta waves, while more advanced systems containing 8 or 16 channels are required for consistent observation of an anterior-posterior gradient. In terms of localization, the 16-channel setup, in principle, was more adept. The 8-channel setup, however, was more effective than the 16-channel setup with regards to displaying the anterior to posterior gradient. Thus, an 8-channel setup is sufficient in an education setting to display these known trends. Modification of the 16-channel setup may provide a better observation of the anterior to posterior gradient.

The fate of spikes in self-limited epilepsy with centrotemporal spikes: Are clinical and baseline EEG features effective?

OBJECTIVE

The aim of this study is to evaluate the effects of clinical and electroencephalographic features on spike reduction with a focus on the first EEG characteristics in self-limited epilepsy with centrotemporal spikes (SeLECTS).

METHODS

This retrospective study was conducted on SeLECTS patients of with at least five years follow-up and at least two EEG recordings in which spike wave indexes (SWI) were calculated.

RESULTS

136 patients were enrolled. Median SWI in the first and last EEGs were 39% (7.6-89%) and 0 (0-112%). Gender, seizure onset age, psychiatric diseases, seizure characteristics (semiology, duration, and relationship to sleep), last EEG time, and spike lateralization in the first EEG did not have a statistically significant effect on the SWI change. Multinomial logistic regression analysis revealed that presence of phase reversal, interhemispheric generalization, and SWI percentage had a significant effect on spike reduction. The frequency of seizures was also significantly decreased in patients with a greater decrease in SWI. Both valproate and levetiracetam were statistically superior in suppressing SWI, with no significant difference between them.

CONCLUSION

Interhemispheric generalization and phase reversal in the first EEG in SeLECTS had negative effects on the spike reduction. The most effective ASMs in reducing spikes were valproate and levetiracetam.

An investigation of the effectiveness of neurofeedback training on motor performance in healthy adults: A systematic review and meta-analysis

Neurofeedback training (NFT) refers to a training where the participants voluntarily aim to manipulate their own brain activity using the sensory feedback abstracted from their brain activity. NFT has attracted attention in the field of motor learning due to its potential as an alternative or additional training method for general physical training. In this study, a systematic review of NFT studies for motor performance improvements in healthy adults and a meta-analysis on the effectiveness of NFT were conducted. A computerized search was performed using the databases Web of Science, Scopus, PubMed, JDreamIII, and Ichushi-Web to identify relevant studies published between January 1st, 1990, and August 3rd, 2021. Thirty-three studies were identified for the qualitative synthesis and 16 randomized controlled trials (374 subjects) for the meta-analysis. The meta-analysis, including all trials found in the search, revealed significant effects of NFT for motor performance improvement examined at the timing after the last NFT session (standardized mean difference = 0.85, 95% CI [0.18-1.51]), but with the existence of publication biases and substantial heterogeneity among the trials. Subsequent meta-regression analysis demonstrated the dose-response gradient between NFTs and motor performance improvements; more than 125 min of cumulative training time may benefit for the subsequent motor performance. For each motor performance measure (e.g., speed, accuracy, and hand dexterity), the effectiveness of NFT remains inconclusive, mainly due to its small sample sizes. More empirical NFT studies for motor performance improvement may be needed to show beneficial effects on motor performance and to safely incorporate NFT into real-world scenarios.

Cross-modal guiding and reweighting network for multi-modal RSVP-based target detection

Rapid Serial Visual Presentation (RSVP) based Brain-Computer Interface (BCI) facilities the high-throughput detection of rare target images by detecting evoked event-related potentials (ERPs). At present, the decoding accuracy of the RSVP-based BCI system limits its practical applications. This study introduces eye movements (gaze and pupil information), referred to as EYE modality, as another useful source of information to combine with EEG-based BCI and forms a novel target detection system to detect target images in RSVP tasks. We performed an RSVP experiment, recorded the EEG signals and eye movements simultaneously during a target detection task, and constructed a multi-modal dataset including 20 subjects. Also, we proposed a cross-modal guiding and fusion network to fully utilize EEG and EYE modalities and fuse them for better RSVP decoding performance. In this network, a two-branch backbone was built to extract features from these two modalities. A Cross-Modal Feature Guiding (CMFG) module was proposed to guide EYE modality features to complement the EEG modality for better feature extraction. A Multi-scale Multi-modal Reweighting (MMR) module was proposed to enhance the multi-modal features by exploring intra- and inter-modal interactions. And, a Dual Activation Fusion (DAF) was proposed to modulate the enhanced multi-modal features for effective fusion. Our proposed network achieved a balanced accuracy of 88.00% (±2.29) on the collected dataset. The ablation studies and visualizations revealed the effectiveness of the proposed modules. This work implies the effectiveness of introducing the EYE modality in RSVP tasks. And, our proposed network is a promising method for RSVP decoding and further improves the performance of RSVP-based target detection systems.

Depression screening using hybrid neural network

Depression is a common cause of increased suicides worldwide, and studies have shown that the number of patients suffering from major depressive disorder (MDD) increased several-fold during the COVID-19 pandemic, highlighting the importance of disease detection and depression management, while increasing the need for effective diagnostic tools. In recent years, machine learning and deep learning methods based on electroencephalography (EEG) have achieved significant results in the field of automatic depression detection. However, most current studies have focused on a small number of EEG signal channels, and experimental data require special processing by professionals. In this study, 128 channels of EEG signals were simply filtered and 24-fold leave-one-out cross-validation experiments were performed using 2DCNN-LSTM classifier, support vector machine, K-nearest neighbor and decision tree. The current results show that the proposed 2DCNN-LSTM model has an average classification accuracy of 95.1% with an AUC of 0.98 for depression detection of 6-second participant EEG signals, and the model is much better than 72.05%, 79.7% and 79.49% for support vector machine, K nearest neighbor and decision tree. In addition, we found that the model achieved a 100% probability of correctly classifying the EEG signals of 300-second participants.

Age-related Deterioration of Alpha Power in Cortical Areas Slowing Motor Command Formation in Healthy Elderly Subjects

BACKGROUND/AIM

The same neural processes may govern older people's motor and cognitive abilities since an inability to switch between actions develops with aging. In this study, a dexterity test was used to measure motor and cognitive perseverance, which required participants to move their fingers fast and correctly on hole boards.

MATERIALS AND METHODS

An electroencephalography (EEG) recording was used to evaluate how healthy young and older adults process brain signals when performing the test.

RESULTS

A significant difference was found between the young and older groups in the average time taken to complete the test, with the older group taking 87.4 s and the young group taking 55.21 s. During motor movement, young participants showed alpha desynchronization over the cortex (Fz, Cz, Oz, Pz, T5, T6, P3, P4) in comparison to the resting state. However, compared to the younger group, no alpha desynchronization was found in the aging group during motor performance. It was noteworthy that alpha power (Pz, P3, and P4) in the parietal cortex was significantly lower in older compared to young adults.

CONCLUSION

Age-related slowdown in motor performance may be caused by deteriorating alpha activity in the parietal cortex, which functions as a sensorimotor interface. This study provides new insights into how perception and action are distributed between brain regions.

Sleep stage classification in EEG signals using the clustering approach based probability distribution features coupled with classification algorithms

Sleep scoring is one of the primary tasks for the classification of sleep stages in Electroencephalogram (EEG) signals. Manual visual scoring of sleep stages is time-consuming as well as being dependent on the experience of a highly qualified sleep expert. This paper aims to address these issues by developing a new method to automatically classify sleep stages in EEG signals. In this research, a robust method has been presented based on the clustering approach, coupled with probability distribution features, to identify six sleep stages with the use of EEG signals. Using this method, each 30-second EEG signal is firstly segmented into small epochs and then each epoch is divided into 60 sub-segments. Each sub-segment is decomposed into five levels by using a discrete wavelet transform (DWT) to obtain the approximation and detailed coefficient. The wavelet coefficient of each level is clustered using the k-means algorithm. Subsequently, features are extracted based on the probability distribution for each wavelet coefficient. The extracted features then are forwarded to the least squares support vector machine classifier (LS-SVM) to identify sleep stages. Comparisons with several existing methods are also made in this study. The proposed method for the classification of the sleep stages achieves an average accuracy rate of 97.4%. It can be an effective tool for sleep stages classification and can be useful for doctors and neurologists for diagnosing sleep disorders.

Heart and brain traumatic stress biomarker analysis with and without machine learning: A scoping review

The enigma of post-traumatic stress disorder (PTSD) is embedded in a complex array of physiological responses to stressful situations that result in disruptions in arousal and cognitions that characterise the psychological disorder. Deciphering these physiological patterns is complex, which has seen the use of machine learning (ML) grow in popularity. However, it is unclear to what extent ML has been used with physiological data, specifically, the electroencephalogram (EEG) and electrocardiogram (ECG) to further understand the physiological responses associated with PTSD. To better understand the use of EEG and ECG biomarkers, with and without ML, a scoping review was undertaken. A total of 124 papers based on adult samples were identified comprising 19 ML studies involving EEG and ECG. A further 21 studies using EEG data, and 84 studies employing ECG meeting all other criteria but not employing ML were included for comparison. Identified studies indicate classical ML methodologies currently dominate EEG and ECG biomarkers research, with derived biomarkers holding clinically relevant diagnostic implications for PTSD. Discussion of the emerging trends, algorithms used and their success is provided, along with areas for future research.

Driver Drowsiness EEG Detection Based on Tree Federated Learning and Interpretable Network

Accurate identification of driver's drowsiness state through Electroencephalogram (EEG) signals can effectively reduce traffic accidents, but EEG signals are usually stored in various clients in the form of small samples. This study attempts to construct an efficient and accurate privacy-preserving drowsiness monitoring system, and proposes a fusion model based on tree Federated Learning (FL) and Convolutional Neural Network (CNN), which can not only identify and explain the driver's drowsiness state, but also integrate the information of different clients under the premise of privacy protection. Each client uses CNN with the Global Average Pooling (GAP) layer and shares model parameters. The tree FL transforms communication relationships into a graph structure, and model parameters are transmitted in parallel along connected branches of the graph. Moreover, the Class Activation Mapping (CAM) is used to find distinctive EEG features for representing specific classes. On EEG data of 11 subjects, it is found that this method has higher average accuracy, F1-score and AUC than the traditional classification method, reaching 73.56%, 73.26% and 78.23%, respectively. Compared with the traditional FL algorithm, this method better protects the driver's privacy and improves communication efficiency.

Classification of Right/Left Hand Motor Imagery by Effective Connectivity Based on Transfer Entropy in Electroencephalogram Signal

Introduction

The right and left-hand motor imagery (MI) analysis based on the electroencephalogram (EEG) signal can directly link the central nervous system to a computer or a device. This study aims to identify a set of robust and nonlinear effective brain connectivity features quantified by transfer entropy (TE) to characterize the relationship between brain regions from EEG signals and create a hierarchical feature selection and classification for discrimination of right and lefthand MI tasks.

Methods

TE is calculated among EEG channels as the distinctive, effective connectivity features. TE is a model-free method that can measure nonlinear effective connectivity and analyze multivariate dependent directed information flow among neural EEG channels. Then four feature subset selection methods namely relief-F, Fisher, Laplacian, and local learningbased clustering (LLCFS) algorithms are used to choose the most significant effective connectivity features and reduce redundant information. Finally, support vector machine (SVM) and linear discriminant analysis (LDA) methods are used for classification.

Results

Results show that the best performance in 29 healthy subjects and 60 trials is achieved using the TE method via the Relief-F algorithm as feature selection and support vector machine (SVM) classification with 91.02% accuracy.

Conclusion

The TE index and a hierarchical feature selection and classification can be useful for the discrimination of right- and left-hand MI tasks from multichannel EEG signals.

Highlights

Effective connectivity features were extracted from electroencephalogram (EEG) to analyze relationships between regions.Four feature selection methods used to select most significant effective features.Support vector machine (SVM) used for discrimination of right and left hand motor imagery (MI) task.

Plain Language Summary

In this study, we investigated brain activity using effective connectivity during MI task based on EEG signals. The motor imagery task can accomplish the same goal as motor execution, since they are both activated by the same brain area. Transfer entropy, coherence, and Granger casualty were employed to extract the features. Differential patterns of activity between the left vs. right MI task showed activity around the motor area rather than other areas. In order to reduce redundant information and select the most significant effective connectivity features, four feature subset selection algorithms are used: Relief-F, Fisher, Laplacian, and learning-based clustering feature selection (LLCFS). Then, support vector machine (SVM) and linear discriminant analysis (LDA) are used to classify left and right hand MI task. Comparison of three different connectivity methods showed that TE index had the highest classification accuracy, and could be useful for the discrimination of right and left hand MI tasks from multichannel EEG signals.

Association between the loudness dependence of auditory evoked potential, serotonergic neurotransmission and treatment outcome in patients with depression

Previous studies have suggested that the loudness dependence of auditory evoked potential (LDAEP) is associated with the effectiveness of antidepressant treatment in patients with major depressive disorders (MDD). Furthermore, both LDAEP and the cerebral serotonin 4 receptor (5-HT₄R) density is inversely related to brain serotonin levels. We included 84 patients with MDD and 22 healthy controls to examined the association between LDAEP and treatment response and its association with cerebral 5-HT₄R density. Participants underwent both EEG and 5-HT₄R neuroimaging with [¹¹C]SB207145 PET. Thirty-nine patients with MDD were re-examined after 8 weeks of treatment with selective serotonin reuptake inhibitors/serotonin noradrenaline reuptake inhibitor (SSRI/SNRI). We found that the cortical source of LDAEP was higher in untreated patients with MDD compared to healthy controls (p=0.03). Prior to SSRI/SNRI treatment, subsequent treatment responders had a negative association between LDAEP and depressive symptoms and a positive association between scalp LDAEP and symptom improvement at week 8. This was not found in source LDAEP. In healthy controls, we found a positive correlation between both scalp and source LDAEP and cerebral 5-HT₄R binding but that was not observed in patients with MDD. We did not see any changes in scalp and source LDAEP in response to SSRI/SNRI treatment. These results support a theoretical framework where both LDAEP and cerebral 5-HT₄R are indices of cerebral 5-HT levels in healthy individuals while this association seems to be disrupted in MDD. The combination of the two biomarkers may be useful for stratifying patients with MDD. Clinical Trials Registration:https://clinicaltrials.gov/ct2/show/NCT02869035?draw=1Registration number: NCT0286903.