Search Results (108)

Electroencephalography (EEG)-based brain–computer interface (BCI) systems pose significant privacy risks, as EEG data remain vulnerable to inference and reconstruction attacks. Conventional privacy-preserving techniques, including data anonymization, encryption, and perturbation, frequently compromise data utility or prove ineffective against advanced adversaries. To address these limitations and balance utility and privacy, we propose a quantum-inspired, differential privacy-based generative adversarial network (Q-DP-GAN). Unlike classical GANs, which lack adaptive privacy mechanisms during training, our method uses quantum-inspired stochasticity to dynamically calibrate noise and the privacy budget. The experimental results demonstrate that Q-DP-GAN is more robust to membership inference and reconstruction attacks than existing approaches. Evaluation on the widely used BCI Competition IV Datasets 2A and 2B indicates that our framework produces high-quality synthetic EEG data while maintaining utility and data confidentiality for BCI classification tasks. Full article

Increasing societal awareness of mental health challenges has significantly reduced stigma surrounding psychological disorders, encouraging greater numbers of individuals to seek professional support, which has placed unprecedented pressure on mental health services, with institutions ranging from educational establishments to emergency services implementing systematic screening protocols to identify individuals requiring intervention. However, the growing demand for rapid, accurate diagnosis continues to strain limited professional resources. Our study introduces an innovative machine learning framework for mental disorder detection using electroencephalography (EEG) signals processed through Welch’s power spectral density estimation. Unlike conventional Fast Fourier Transform (FFT) approaches, our method generates refined two-dimensional spectrograms capturing brain wave amplitudes (in dB) alongside precise peak frequency identification. This computationally efficient periodogram variant enables robust feature extraction suitable for real-time diagnostic applications while reducing model training overhead. Preliminary analysis demonstrates the Welch Transform’s superior signal characterization compared to standard FFT periodograms, revealing distinct neurophysiological patterns associated with various mental health conditions. The approach maintains high computational efficiency, supporting potential deployment in clinical screening environments. Full article

This study quantitatively evaluates the performance of a non-invasive hybrid brain–computer interface (BCI) compared to a conventional mouse in pointing (point-and-click) tasks. A commercial wearable BCI (Brainfingers), based on electromyography (EMG) and electrooculography (EOG) signals with low-level electroencephalography (EEG) components, was assessed against a Microsoft Optical Mouse using ISO/TS 9241-411-based one-dimensional (1D) and two-dimensional (2D) target acquisition tasks. Pointer coordinates were recorded and analyzed using Fitts’ law metrics. A total of 48 non-disabled participants completed the experiments. The results reveal significant performance differences between the two input devices. The BCI device exhibits substantially lower performance than the mouse across the reported Fitts’ law measures. Mean throughput was 0.35 bits/s for the BCI and 6.03 bits/s for the mouse in the 1D tests and 0.43 bits/s for the BCI and 5.17 bits/s for the mouse in the 2D tests. Despite the BCI’s low performance and although the present experiments involved non-disabled participants, the findings, considered alongside the prior literature on Brainfingers and non-invasive BCIs for computer access, suggest that the device may still have assistive technology value for users with severe motor impairments. Full article

Neonates undergoing surgery for congenital heart disease (CHD) are at high risk for brain function impairment. Reliable early predictors of postoperative neurological complications are lacking. We examined a retrospective cohort of 55 surgically treated CHD neonates systematically monitored by concomitant conventional electroencephalography (cEEG) and amplitude-integrated EEG (aEEG). Neonates underwent cEEG/aEEG at three time points: T0 (preoperative, duration: 90–120 min); T1 (24–48 h after cardiac surgery, duration: ≥11 h); and T2 (7–10 days post-surgery, duration: 90–120 min). For each patient, aEEG background activity was evaluated and scored, and clinical and surgical data were retrieved to establish short-term post-surgical outcomes. Patients with normal T0 monitoring had significantly higher aEEG bandwidths in T1. A lower Aristotle basic score was associated with an improvement in aEEG at T1. Inversely, a narrower aEEG bandwidth in T1 was associated with post-surgical neurological deterioration. The aEEG bandwidth accurately predicted short-term neurological outcome; in particular, a minimal aEEG amplitude above 17.5 µV excluded poor neurological outcome with a negative predictive value of 81.48%. Our results demonstrated that aEEG bandwidth and trend dynamics may be associated with surgical complexity and neurological outcomes. aEEG background trend monitoring may provide relevant prognostic information on neurological outcomes in surgically treated CHD neonates. Full article

Electroencephalography (EEG) provides a non-invasive and high-temporal-resolution modality for decoding cognitive states, but high-density recordings remain challenging for Transformer-based models because self-attention scales quadratically with the number of channels. In addition, conventional Euclidean representations do not fully capture the intrinsic geometry of EEG covariance features, which may limit robustness in cross-subject settings. To address these issues, we propose EEG-RCformer, a Riemannian geometry-informed channel clustering Transformer for EEG decoding. The model first computes per-channel symmetric positive definite (SPD) covariance matrices from windowed EEG features and uses the affine-invariant Riemannian metric (AIRM) to identify trial-specific functional hubs. These hubs are then integrated with capacity-constrained spatial clustering to generate anatomically plausible and computationally efficient channel groups, which are encoded as tokens for a Transformer classifier. We evaluated EEG-RCformer on the MODMA and SEED datasets under both subject-dependent and -independent paradigms, achieving area under the curve (AUC) values of 0.9802 and 0.7154 on MODMA and 0.8541 and 0.8011 on SEED, respectively. Paired statistical tests further showed significant gains for MODMA in both the subject-dependent and -independent settings and for SEED in the subject-dependent setting, while SEED still showed a positive but non-significant mean improvement in the subject-independent setting. Full article

Introduction: Caffeine is the most widely consumed psychoactive stimulant worldwide and acts primarily through antagonism of adenosine A1 and A2A receptors, thereby reducing sleep pressure and promoting wakefulness. Although its alerting and performance-enhancing effects are well established, its influence on sleep-related electroencephalography (EEG) has been investigated across diverse paradigms with substantial methodological heterogeneity. This systematic and mechanistic review aimed to synthesize human evidence on how caffeine affects sleep architecture, quantitative sleep EEG, and neurophysiological markers of sleep homeostasis, and to interpret these findings within current models of adenosine-mediated sleep–wake regulation. Materials and Methods: A systematic search of PubMed/MEDLINE, Web of Science, Scopus, Embase, PsycINFO, ResearchGate, and Google Scholar was conducted for studies published between January 1980 and January 2026, with the final search performed on 10 January 2026. Eligible studies were original human investigations examining caffeine exposure or administration and reporting sleep-related EEG outcomes, including polysomnographic sleep staging, spectral EEG analyses, or other EEG-derived sleep metrics. Two reviewers independently screened records and assessed eligibility, with disagreements resolved by consensus. Data on study design, participant characteristics, caffeine interventions, EEG methodology, and outcomes were extracted using a predefined form. Risk of bias was evaluated using the RoB 2 and ROBINS-I tools. Owing to marked heterogeneity across studies, findings were synthesized narratively within a mechanistic interpretive framework. Results: Thirty-two studies were included. Across highly heterogeneous paradigms—including acute bedtime or evening dosing, daytime or repeated caffeine use before nocturnal sleep, administration during prolonged wakefulness followed by recovery sleep, withdrawal protocols, and ambulatory/home EEG monitoring—the most consistent finding was suppression of low-frequency NREM EEG activity, particularly slow-wave activity and the lowest delta frequencies. Caffeine frequently increased faster EEG activity, including sigma/spindle and beta ranges, producing a lighter, more aroused, and more wake-like sleep EEG profile. These effects were especially prominent during early-night NREM sleep and in recovery sleep after sleep deprivation, where caffeine attenuated the expected homeostatic rebound in low-frequency power. REM-related effects were less consistent, but some studies reported delayed REM timing and subtler alterations in REM EEG. Emerging evidence further suggests that caffeine increases EEG complexity and shifts sleep dynamics toward a more excitation-dominant state. Several studies indicated that quantitative EEG measures were more sensitive than conventional sleep-stage variables in detecting caffeine-related sleep disruption. Dose, timing, habitual caffeine use, withdrawal state, age, circadian context, and adenosinergic genetic variation, particularly involving ADORA2A, moderated the magnitude of effects. We also highlighted the connection between current results and sports and sports science. Conclusions: Caffeine reliably alters the neurophysiological architecture of human sleep in a direction consistent with reduced sleep depth and weakened homeostatic recovery. The overall evidence supports a mechanistic model centered on adenosine receptor antagonism, attenuation of sleep-pressure build-up and expression, and a shift toward greater cortical arousal during sleep. Sleep EEG appears to be a sensitive marker of these effects, often revealing physiological disruption even when conventional sleep architecture changes are modest. Future research should prioritize larger and more diverse samples, pharmacokinetic and pharmacogenetic characterization, and ecologically valid high-resolution sleep monitoring to clarify the real-world and functional consequences of caffeine-induced EEG changes. Full article

Background/Objectives: Advanced childhood cerebral X-linked adrenoleukodystrophy (cALD) is traditionally regarded as an irreversible terminal phase of neurodegeneration driven by inflammatory demyelination and axonal loss. Experimental evidence indicates that endogenous bioelectrical fields regulate central nervous system organisation, raising the possibility that functional network collapse in cALD may be biologically modifiable, even in the presence of persistent structural damage. This study examined whether longitudinal modulation of endogenous bioelectrical network organisation is associated with sustained clinical and neurophysiological stabilisation in advanced cALD. Methods: We performed a longitudinal observational analysis of two paediatric patients with advanced childhood cerebral X-linked adrenoleukodystrophy undergoing repeated neuroregenerative treatment cycles. Standardised scalp electroencephalography was recorded during spontaneous wakefulness and repeated over months under comparable vigilance conditions. Multimodal analysis included conventional EEG, quantitative EEG, independent component analysis, and standardised low-resolution electromagnetic tomography (sLORETA). Clinical function was assessed using validated measures of consciousness, swallowing, and voluntary motor behaviour. Results: Across patients, longitudinal recordings demonstrated sustained stabilisation of consciousness, swallowing, and voluntary motor function, accompanied by reproducible reorganisation of pathological brain rhythms. Delta and theta oscillations showed a consistent topographical redistribution from limbic–frontoinsular networks towards sensorimotor and parietal integrative cortices. These changes were observed across modalities and timepoints and are unlikely to reflect spontaneous fluctuation, delayed effects of haematopoietic stem cell transplantation, or state-dependent EEG variation. Conclusions: Advanced childhood cerebral X-linked adrenoleukodystrophy is associated with disorganisation of endogenous bioelectrical network activity. In this longitudinal analysis, large-scale network reorganisation was temporally associated with sustained clinical stabilisation, supporting a view of late-stage cALD as a dynamic disorder of network-level vulnerability, rather than a fixed terminal state. Full article

Rapid post-disaster building damage assessment requires recognizing explicit structural failures and interpreting implicit situational cues in visually complex scenes. Whereas conventional automated methods are often confined to detecting explicit damage patterns, human perception naturally integrates both types of information into a holistic risk judgment. This study presents an exploratory investigation into the neural signatures underlying this integrated judgment process using electroencephalography. A modified paradigm was employed to probe the cognitive dynamics of risk evaluation in participants with civil engineering backgrounds. Although participants were instructed only to identify damaged buildings without explicit severity grading, event-related potential analysis revealed systematic, graded neural responses that scaled with damage severity. This suggests that the brain encodes damage-related information not as a binary state but as a continuous spectrum of perceived risk, implicitly processing severity, even in the absence of explicit instructions. Furthermore, single-trial analysis demonstrated that time-domain features contain robust discriminative information, verifying the feasibility of decoding these latent judgments from brain activity. These findings provide a physiological basis for developing future cognition-informed algorithms and human-in-the-loop frameworks, bridging the semantic gap to enhance the reliability of automated disaster assessment. Full article

Point-of-care electroencephalography (POC-EEG) has emerged as a practical tool for the rapid detection of non-convulsive seizures (NCS) and non-convulsive status epilepticus (NCSE) in acute neurological settings where access to conventional EEG is often delayed. This narrative review synthesizes current evidence on the clinical applications, tech-no-logical evolution, and limitations of POC-EEG systems across adult and pediatric populations. Available data suggest that POC-EEG is associated with earlier seizure identification, more timely antiseizure treatment decisions, and reduced dependence on inter-facility transfers in selected healthcare settings. Beyond seizure detection, POC-EEG has shown potential utility in the assessment of acute encephalopathy due to conditions such as stroke, traumatic brain injury, delirium, and post-cardiac arrest states. Recent advances in device portability and artificial intelligence-assisted interpretation have expanded accessibility, enabling use by non-specialist clinicians; however, reduced spatial resolution, artifact susceptibility, and variable performance in focal or low-burden epileptiform activity remain important limitations. Automated detection algorithms show high accuracy for sustained seizure burden but require cautious interpretation and further prospective validation. Ethical and health-system considerations, including equitable access, diagnostic stewardship, and data governance, are increasingly relevant as adoption grows. Overall, POC-EEG represents a promising adjunct to conventional EEG that may improve early diagnostic workflows in acute neurological care, while definitive impacts on long-term outcomes warrant further study. Full article

(1) Background: Electroencephalography (EEG) is widely used to monitor the depth of anesthesia; however, conventional Fourier-based analyses are limited in their ability to characterize non-stationary anesthetic-induced EEG dynamics. In this study, we investigated the utility of singular spectrum analysis (SSA) combined with the Hilbert transform for extracting physiologically meaningful EEG features under sevoflurane general anesthesia. (2) Methods: Frontal EEG data from ten patients undergoing sevoflurane anesthesia were analyzed from the maintenance phase through emergence. Using SSA, short EEG segments were decomposed into six intrinsic mode functions (IMFs) without pre-specified basis functions or frequency bands. Hilbert spectral analysis was applied to each IMF to obtain instantaneous frequency and amplitude characteristics. (3) Results: The SSA-based decomposition clearly captured phase-dependent EEG changes, including α spindle activity during maintenance and increasing high-frequency components preceding emergence. Multiple linear regression models incorporating IMF center frequencies and total power demonstrated strong correlations with the bispectral index (BIS), achieving high predictive accuracy (R² = 0.88, MAE < 4). Compared with conventional spectral approaches, SSA provided superior temporal resolution and stable feature extraction for non-stationary EEG signals. (4) Conclusions: These findings indicate that SSA combined with Hilbert analysis is a robust framework for quantitative EEG analysis during general anesthesia and may enhance real-time, individualized assessments of anesthetic depth. Full article

In this study, we introduce a novel, functionally driven method for model pruning guided by sensitivity analysis. Conventional model compression techniques often rely on proxy metrics, such as weight magnitude, which may not accurately reflect a component’s true functional importance. The proposed method directly assesses this by systematically injecting controlled noise into network layers and measuring the resultant perturbation on inference output. Components exhibiting low sensitivity to this noise are identified as functionally redundant and are pruned. We validated the method on EEGNet, a compact convolutional neural network, using the MNE Sample Event-Related Potential (ERP) dataset, a widely used benchmark for electroencephalography classification. After training the baseline model, we generated a sensitivity profile by quantifying how noise injection at different layers impacts predictive accuracy. This profile then guided targeted pruning of less influential convolutional kernels and weights. Experimental results demonstrate the method’s efficacy, achieving a significant reduction in both parameter count and computational complexity. Crucially, the pruned model retains classification accuracy nearly identical to the original, heavyweight EEGNet. This confirms that sensitivity-guided pruning effectively removes redundancy without degrading performance. In conclusion, our noise injection framework provides a more direct and interpretable criterion for neural network simplification. By linking component pruning to functional impact, our method enables a more precise and efficient model reduction than traditional heuristic-based approaches. The method developed presents a practical pathway toward developing lightweight, accurate, and low-latency models essential for real-world neuro-computational applications. Future work will focus on automating the pruning pipeline and extending the framework’s applicability to diverse neural architectures. Full article

Electroencephalography-based motor imagery (EEG-MI) classification is a cornerstone of Brain–Computer Interface (BCI) systems, enabling the identification of motor intentions by decoding neural patterns within EEG signals. However, conventional methods, predominantly reliant on convolutional neural networks (CNNs), are proficient at extracting local temporal features but struggle to capture long-range dependencies and global contextual information. To address this limitation, we propose a Triple-path Heterogeneous Feature Collaboration Network (TPHFC-Net), which synergistically integrates three distinct temporal modeling pathways: a multi-scale Temporal Convolutional Network (TCN) to capture fine-grained local dynamics, a Transformer branch to model global dependencies via multi-head self-attention, and a Long Short-Term Memory (LSTM) network to track sequential state evolution. These heterogeneous features are subsequently fused adaptively by a dynamic gating mechanism. In addition, the model’s robustness and discriminative power are further augmented by a lightweight front-end denoising diffusion model for enhanced noisy feature representation and a back-end prototype attention mechanism to bolster the inter-class separability of non-stationary EEG features. Extensive experiments on the BCI Competition IV-2a and IV-2b datasets validate the superiority of the proposed model, achieving mean classification accuracies of 82.45% and 89.49%, respectively, on the subject-dependent MI task and significantly outperforming existing mainstream baselines. Full article

Smart devices and multimodal biosignal systems, including electroencephalography (EEG/MEG), ECG-derived heart rate variability (HRV), and electromyography (EMG), increasingly supported by artificial intelligence (AI), are being explored to improve the assessment and longitudinal monitoring of mental health conditions. Despite rapid growth, the available evidence remains heterogeneous, and clinical translation is limited by variability in acquisition protocols, analytical pipelines, and validation quality. This systematic review synthesizes current applications, signal-processing approaches, and methodological limitations of biosignal-based smart systems for mental health monitoring. Methods: A PRISMA 2020-guided systematic review was conducted across PubMed/MEDLINE, Scopus, the Web of Science Core Collection, IEEE Xplore, and the ACM Digital Library for studies published between 2013 and 2026. Eligible records reported human applications of wearable/smart devices or multimodal biosignals (e.g., EEG/MEG, ECG/HRV, EMG, EDA/GSR, and sleep/activity) for the detection, monitoring, or management of mental health outcomes. The reviewed literature after predefined inclusion/exclusion criteria clustered into six themes: depression detection and monitoring (37%), stress/anxiety management (18%), post-traumatic stress disorder (PTSD)/trauma (5%), technological innovations for monitoring (25%), brain-state-dependent stimulation/interventions (3%), and socioeconomic context (7%). Across modalities, common analytical pipelines included artifact suppression, feature extraction (time/frequency/nonlinear indices such as entropy and complexity), and machine learning/deep learning models (e.g., SVM, random forests, CNNs, and transformers) for classification or prediction. However, 67% of studies involved sample sizes below 100 participants, limited ecological validity, and lacked external validation; heterogeneity in protocols and outcomes constrained comparability. Conclusions: Overall, multimodal systems demonstrate strong potential to augment conventional mental health assessment, particularly via wearable cardiac metrics and passive sensing approaches, but current evidence is dominated by proof-of-concept studies. Future work should prioritize standardized reporting, rigorous validation in diverse real-world cohorts, transparent model evaluations, and ethics-by-design principles (privacy, fairness, and clinical governance) to support translation into practice. Full article

Background/Objectives: Cognitive decline and compromised attention control serve as early indicators of neurodysfunction that manifest as broader psychopathological symptoms, yet conventional mental health assessment relies predominantly on subjective self-report measures lacking objectivity and temporal granularity. We propose Neurosense, an AI-driven brain health assessment framework using electroencephalography (EEG) to non-invasively capture neural dynamics. Methods: Our Dual-path Spatio-Temporal Adaptive Gated Encoder (D-STAGE) architecture processes temporal and spatial EEG features in parallel through Transformer-based and graph convolutional pathways, integrating them via adaptive gating mechanisms. We introduce a two-stage paradigm: first training on cognitive task EEG for reaction time prediction to acquire cognitive performance-related representations, then featuring parameter-efficient adapter-based transfer learning to estimate p-factor—a transdiagnostic psychopathology dimension. The adapter-based transfer achieves competitive performance using only 1.7% of parameters required for full fine-tuning. Results: The model achieves effective reaction time prediction from EEG signals. Transfer learning from cognitive tasks to mental health assessment demonstrates that cognitive efficiency representations can be adapted for p-factor prediction, outperforming direct training approaches while maintaining parameter efficiency. Conclusions: The Neurosense framework reveals hierarchical relationships between neural dynamics, cognitive efficiency, and mental health dimensions, establishing foundations for a promising computational framework for mental health assessment applications. Full article

Background/Objectives: Sleep stage classification is crucial for assessing sleep quality and diagnosing related disorders. Electroencephalography (EEG) is currently recognized as a primary method for sleep stage classification. High-performance automatic sleep staging methods based on EEG leverage the powerful contextual modeling capabilities of Transformer Encoder architectures. However, the global self-attention mechanism in Transformers incurs significant computational overhead, substantially hindering the training and inference efficiency of automatic sleep staging algorithms. Methods: To address these issues, we introduce an end-to-end framework for automatic sleep stage classification using single-channel EEG: SleepMFormer. At the algorithmic level, SleepMFormer adopts a task-driven simplification of the Transformer encoder to improve attention efficiency while preserving sequence modeling capability. At the training level, supervised contrastive learning is incorporated as an auxiliary strategy to enhance representation robustness. From an engineering perspective, these design choices enable efficient training and inference under resource-constrained settings. Results: When integrated with the SleePyCo backbone, the proposed framework achieves competitive performance on three widely used public datasets: Sleep-EDF, PhysioNet, and SHHS. Notably, SleepMFormer reduces training and inference time by up to 33% compared to conventional self-attention-based models. To further validate the generalizability of MaxFormer, we conduct additional experiments using DeepSleepNet and TinySleepNet as alternative feature extractors. Experimental results demonstrate that MaxFormer consistently maintains performance across different model architectures. Conclusions: Overall, SleepMFormer introduces an efficient and practical framework for automatic sleep staging, demonstrating strong potential for related clinical applications. Full article