The success of deep learning (DL) methods in the Brain-Computer Interfaces (BCI) field for classification electroencephalographic (EEG) recordings has been restricted by lack large datasets. Privacy concerns associated with EEG signals limit possibility constructing a EEG-BCI dataset conglomeration multiple small ones jointly training machine models. Hence, this paper, we propose novel privacy-preserving DL architecture named federated transfer (FTL) that is based on framework. Working...

Electroencephalography (EEG) classification techniques have been widely studied for human behavior and emotion recognition tasks. But it is still a challenging issue since the data may vary from subject to subject, change over time same maybe heterogeneous. Recent years, increasing privacy-preserving demands poses new challenges this task. The heterogeneity, as well privacy constraint of EEG data, not concerned in previous studies. To fill gap, paper, we propose heterogeneous federated...

The neurological disorder epilepsy causes substantial problems to the patients with uncontrolled seizures or even sudden deaths. Accurate detection and prediction of epileptic will significantly improve life quality patients. Various feature extraction algorithms were proposed describe EEG signals in frequency time domains. Both invasive intracranial non-invasive scalp have been screened for seizure patterns. This study extracted a comprehensive list 24 types from found 170 out 2794 features...

Background The number of patients with Alzheimer’s disease (AD) worldwide is increasing yearly, but the existing treatment methods have poor efficacy. Transcranial alternating current stimulation (tACS) a new for AD, offline effect tACS insufficient. To prolong effect, we designed to combine sound maintain long-term post-effect. Materials and explore safety effectiveness combined its impact on cognition AD patients. This trial will recruit 87 mild moderate AD. All were randomly divided into...

In recent years, with the shift of focus in metaverse research toward content exchange and social interaction, breaking through current bottleneck audio-visual media interaction has become an urgent issue. The use brain-machine interfaces for sensory simulation is one proposed solutions. Currently, have demonstrated irreplaceable potential as physiological signal acquisition tools various fields within metaverse. This study explores three application scenarios: generative art metaverse,...

Objective. The number of items a P300-based brain-computer interface (BCI) should be adjustable in accordance with the requirements specific tasks. To address this issue, we propose novel task-oriented optimal approach aimed at increasing performance general P300 BCIs different numbers items. Approach. First, proposed stimulus presentation variable dimensions (VD) paradigm as generalization conventional single-character (SC) and row-column (RC) paradigms. Furthermore, an embedding design was...

ABSTRACT Under the dual influence of power system transition to integrated energy and evolution cyberattack technology, a correlation feature‐multilabel cascade boosted forest based false data injection attack localization detection method is proposed for new accurately locate attacked position grid in response stealthy (FDIA). Considering FDIA principle characteristics system, as well fact that contains large amount measurement variable operation states, enhances fitting ability multi‐label...

The Scalp Electroencephalogram (EEG) signal of epileptic patients often contains Interictal Epileptiform Discharges (IED) during the period seizures. Detection IEDs is significant for diagnosis epilepsy and prediction In this paper, we proposed a graph convolutional network bi-directional LSTM co-embedded broad learning system to detect IEDS. Here, represent EEG as utilize Graph Convolutional Networks (GCN) extract contextual features. addition, also adopted extracting temporal feature from...

As an effective alternative to deep neural networks, broad learning system (BLS) has attracted more attention due its efficient and outstanding performance shorter training process in classification regression tasks. Nevertheless, the of BLS will not continue increase, but even decrease, as number nodes reaches saturation point continues increase. In addition, previous research on networks usually ignored reason for good generalization networks. To solve these problems, this article first...

A seizure is a neurological disorder caused by abnormal neuronal discharges in the brain, which severely reduces quality of life patients and often endangers their lives. Automatic detection an important research area treatment prerequisite for intervention. Deep learning has been widely used automatic seizures, many related works decomposed electroencephalogram (EEG) raw signal with time window to obtain EEG slices, then performed feature extraction on represented obtained features as input...

Transcranial alternating current stimulation (tACS) is a relatively new non-invasive brain electrical method for the treatment of patients with Alzheimer's disease (AD), but it has poor offline effects. Therefore, we applied combined to observe effect on cognitive function AD. Here, describe clinical results case in which tACS sound was treat moderate The patient 73-year-old woman 2-year history persistent deterioration despite administration Aricept and Sodium Oligomannate. received...

This study proposes several categories of stimulus types under the visual oddball speller paradigm with application to brain-computer interface (BCI). Motion (including translation and rotation), zoom in/out, pattern rotation sharpening are tested analysed. Results show that zooming type could obtain comparative or higher accuracies as well information transfer rates compared classic color type. Spatio-temporal distributions weighted event related potentials evoked by different respective...

As Convolutional Neural Networks continue to produce state of the art results, more types data are being used see results that would be produced. Using heart rate was collected using sensors from various subjects who consumed alcohol, we converted it 1D waveform into a set spectrograms. The spectrograms were fed two pretrained CNNs, CaffeNet and AlexNet, determine whether or not given spectrogram an instance alcohol consumption. 80 training images (40 positive, 40 negative) 20 test (10 10...

EEG-based emotion recognition has become an important part of human–computer interaction. To solve the problem that single-modal features are not complete enough, in this paper, we propose a multimodal method based on attention recurrent graph convolutional neural network, which is represented by Mul-AT-RGCN. The explores relationship between multiple-modal feature channels EEG and peripheral physiological signals, converts one-dimensional sequence into two-dimensional map for modeling, then...

Drowsy driving has a crucial influence on safety, creating an urgent demand for driver drowsiness detection. Electroencephalogram (EEG) signal can accurately reflect the mental fatigue state and thus been widely studied in monitoring. However, raw EEG data is inherently noisy redundant, which neglected by existing works that just use single-channel or full-head channel model training, resulting limited performance of In this paper, we are first to propose Interpretability-guided Channel...

Anesthesia signal monitoring is a very important indicator in surgery, and the effective of anesthesia depth has been goal anesthesiologists biomedical engineering experts recent decades. First, wavelet transform method used to analyze EEG signals, extracted features are clustered by classifier estimate anesthesia. Second, characteristics eigenvectors constructed singular value decomposition based on coefficients. The extraction extracts mid-latency auditory evoked under Finally, this paper...

Objective: To investigate the spontaneous brain activity changes with cognitive disorders of traumatic injury in frontal lobe using resting-state fMRI and regional homogeneity (ReHo).Methods: Thirteen injuries cognition impairment were sampled as group (TBIs) fourteen healthy persons normal control group(NCs).General was assessed through Mini-Mental State Examination(MMSE).Resting-state T1-weighted imaging data then collected.The ReHo maps obtained from fMRI, two-sample t-test performed...

The quality of sleep has a great relationship with health. result stage classification is an important indicator to measure the sleep. It was found that symbolic transfer entropy about wake and first non-rapid eye movement reflect on changes stage. And it confirmed by T test multi-samples experiments. can apply into automatic classification. By Multi-parameter analysis could achieve higher accuracy