Résultats de recherche pour « EEG »

Dementia (DEM) is a growing global health challenge, underscoring the need for early and accurate diagnosis. Electroencephalography (EEG) provides a non-invasive window into brain activity, but conventional methods struggle to capture its transient complexity. We present the \textbf{EEG Microstate Analysis Framework (EEG-MSAF)}, an end-to-end pipeline that leverages EEG microstates discrete, quasi-stable topographies to identify DEM-related biomarkers and distinguish DEM, mild cognitive impairment (MCI), and normal cognition (NC). EEG-MSAF comprises three stages: (1) automated microstate feature extraction, (2) classification with machine learning (ML), and (3) feature ranking using Shapley Additive Explanations (SHAP) to highlight key biomarkers. We evaluate on two EEG datasets: the public Chung-Ang University EEG (CAUEEG) dataset and a clinical cohort from Thessaloniki Hospital. Our framework demonstrates strong performance and generalizability. On CAUEEG, EEG-MSAF-SVM achieves \textbf{89\% $\pm$ 0.01 accuracy}, surpassing the deep learning baseline CEEDNET by \textbf{19.3\%}. On the Thessaloniki dataset, it reaches \textbf{95\% $\pm$ 0.01 accuracy}, comparable to EEGConvNeXt. SHAP analysis identifies mean correlation and occurrence as the most informative metrics: disruption of microstate C (salience/attention network) dominates DEM prediction, while microstate F, a novel default-mode pattern, emerges as a key early biomarker for both MCI and DEM. By combining accuracy, generalizability, and interpretability, EEG-MSAF advances EEG-based dementia diagnosis and sheds light on brain dynamics across the cognitive spectrum.

Recent EEG-to-image retrieval models have achieved strong performance in identifying viewed images from semantically diverse candidates. Yet such success does not reveal what visual information supports the match. A model may readily identify a cheetah among tools, plants, and vehicles, but can it still distinguish the viewed cheetah from the same scene with the cheetah replaced by a dog? Motivated by this question, we introduce EEG-EditBench, a diagnostic benchmark that examines this question through controlled edits of object identity, attributes, background, and object presence. Built from the 200 THINGS-EEG2 test images, EEG-EditBench contains 2,137 quality-controlled edits and evaluates eight representative EEG visual decoding models. Our results show that strong standard retrieval does not consistently transfer to edit-based evaluation, with fine-grained attribute changes presenting the greatest challenge. EEG-EditBench reveals model behavior hidden by aggregate retrieval accuracy and provides a controlled basis for studying what visual information EEG-image models preserve. The code and complete dataset are publicly available.

Electroencephalography (EEG) supports a variety of brain-computer interface (BCI) tasks ranging from brain-state monitoring to human-LLM interactions. EEG foundation models are emerging, but evaluation remains fragmented due to heterogeneous datasets and nconsistent task protocols. Here, we introduce OmniEEG-Bench, a unified benchmark and downstream task roadmap for EEG foundation models (FMs). It organizes evaluation of EEG FMs into six task families spanning (i) signal reliability, (ii) biometrics and disease, (iii) consciousness and state, (iv) cognition and emotion, (v) naturalistic stimulus decoding, and (vi) motor and interaction, introducing a new generation of tasks not systematically benchmarked in prior EEG FM work. OmniEEG-Bench standardizes model deployment, task definitions, and metrics through a task-card specification, and unifies 54 EEG datasets with consistent evaluation protocols. We benchmark 10 representative EEG foundation models and report a leaderboard that covers diverse evaluation settings. Both pretraining dataset diversity and model size are significantly associated with better average ranks across datasets, revealing scaling-law behavior in EEG foundation models (Figure 1). These results suggest that scaling EEG foundation models requires not only larger architectures but also broader and more diverse pretraining data. The benchmark code is available at https://github.com/ncclab-sustech/omni-eegbench.git.

The development of generalizable electroencephalography (EEG) decoding models is essential for robust brain-computer interfaces (BCI) and objective neural biomarkers in mental health. Conventional approaches have been hindered by poor cross-subject and cross-task generalization, owing to high inter-subject variability and non-stationary neural signals. We address this challenge with a zero-shot cross-subject decoding framework on the large-scale Healthy Brain Network dataset, benchmarking a convolutional neural network baseline, a hybrid LSTM, and a Transformer-based foundation model. To adapt the Transformer for regression while averting catastrophic forgetting, we propose a novel progressive unfreezing strategy. The baseline yielded an nRMSE of 0.9991, whereas our fine-tuned Transformer achieved 0.9799 on unseen subjects. This work advances scalable, calibration-free EEG decoding for computational psychiatry and behavioral prediction.

The true frequency range of the EEG is much broader than it has been assumed and taught for decades. The EEG apparatuses with inkwriting pens recording on paper are incapable of giving us trustworthy tracings beyond 80/sec. With the introduction of digital EEG machines, the exploration of the 60 to 1000 Hz range has already begun in the past few years (but, strangely enough, had been in use during the pioneer age when short photographic EEG recordings were made). The new wave of ultrafast recording began in the domain of somatosensory evoked potentials (SSEP). In the field of EEG (strictly speaking), research work started very recently. Ultrafast EEG activity promises new insights into the electrophysiological basis of epileptic phenomena. Activities from 150- to 500/sec have been noted in recent studies (including personal work). Faster frequencies (500–1000/sec) are likely to play a major role in the electrophysiology of neurocognition and motor initiation. Such EEG-based neurocognitive studies will provide us with in-real-time data and thus outperform PET scanning and functional MRI. Even ultrafast EEG activity has its limitation, which appears to lie around 1000/sec. Faster frequencies (1000–3000 Hz) — recorded mainly with cathode ray oscillography — are probably incompatible with the shortest duration of true field potentials and might be nothing but “neuronal chatter.”

Concurrency of transcranial magnetic stimulation with electroencephalography (TMS-EEG) technique is a powerful and challenging methodology for basic research and clinical applications. Aspects considered in experiments for effective TMS-EEG recordings and analysis, including artifact management, data analysis and interpretation and protocols. mini review offers an extensive insight of TMS-EEG methodology in experimental and computational procedures. Case study aims to leverage an openly available, high-quality EEG dataset to delve into the alterations in cortical activity. By applying Intermittent theta-burst stimulation (iTBS) and continuous theta-burst stimulation (cTBS) to the left dorsolateral prefrontal cortex (DLPFC) in healthy individuals, we observe changes in oscillatory patterns within the EEG data. The dataset includes meticulously extracted resting-state EEG recordings, TMS-evoked potential data, and MRI scans. To process these data, we utilized Brainstorm, an open-source Matlab application, which facilitated noise reduction through independent component analysis and signal-space projection techniques. It allowed us to identify, visualize, and analyze TMS-evoked potentials (TEPs) and TMS-induced oscillations (TIOs). In addition, the study presents detailed plots of resting-state EEG power, local mean field power (LMFP), TMS-related spectral perturbation (TSRP), and inter-trial phase clustering (ITPC). Paired t-tests and cluster-based permutation tests have been performed for statistical analysis. The wealth and quality of this dataset make it ideal for examining the neuromodulatory impact of TBS on the prefrontal cortex. Brainstorm's extensive feature set greatly supports the exploration of such neurological data. Future research directions could concentrate on conducting source localization analyses and comparative group studies.