eeg

BridgeRouter 将多桥跨模态迁移建模为实例级算法选择问题，通过增益感知的轻量路由器在冻结的 EEG 基础模型桥路径间进行推理时路由，在 FOG 和 ISRUC 数据集上显著优于单一最佳桥路径。

A controlled evaluation of five public EEG foundation models plus an in-house JEPA-style model, showing that random-initialisation, band-power baselines, and validation-selected probe layers substantially alter reported pretraining gains on sleep, emotion, and motor imagery benchmarks.

论文提出 FOCUS——一个基于 Mamba-2 的超轻量 EEG 认知负荷分类模型，采用固定 FFT 频带分解与权值共享编码，在仅 0.04M 参数下于 11 人试点实验中达到 74.75% 准确率与 0.08ms 前向延迟。

这篇论文将音乐愉悦度预测构建为回归问题，利用MUSIN-G数据集的128通道EEG数据，比较了六种深度学习模型（DeepConvNet、CNN-LSTM、GNN-CNN、TCN等），并通过可解释性分析识别影响预测的关键时间片段。

该论文提出一个对比学习框架，将EEG信号与冻结的预训练文本编码器的嵌入空间对齐，用于解码想象言语，并在Chisco数据集上验证了其在固定意图解码、相似度检索和语义结构探针方面的能力。

MontageFlow 是一个流匹配模型，通过将输出限制在满足 EEG 导联（montage）线性约束的子空间内，生成物理上有效的高保真临床 EEG 信号，在三个大规模基准上将 TS-FID 提升了 1.6–3.6 倍。

研究人员开发了基于 ConvNeXt 深度学习模型的方法，用于检测新生儿脑电图中的低电压间期（LVI），其性能达到专家级水平（AUC 0.996），并证明 LVI 可作为新生儿脑病的稳健临床生物标志物，与 EEG 背景活动、HIE 严重程度及癫痫风险强相关。

该论文提出 Language as Supervision (LaS) 方法，将神经特征与自然语言类别描述对齐，用于脑机接口解码，在语音解码上较传统类别标签监督基线提升最多 17.15 个百分点，并支持未见类别泛化。

该研究使用在 Temple University EEG 语料上预训练的轻量级 CNN–Transformer 基础模型提取个性化潜在动力学，发现癫痫与非癫痫患者即使预测拟合度几乎相同，其学到的潜在依赖结构却存在显著差异，强调个性化临床模型需同时评估预测效度与内部动力学结构。

SPERA 是一种采用 JEPA 潜空间预测架构的 EEG 基础模型，引入 Legendre 多项式空间先验与分解式时空注意力，在约 8 万小时、29,048 名受试者的 EEG 数据上预训练，并在九项下游任务上取得最高平均平衡准确率。

The paper studies zero-shot cross-subject continuous valence-arousal regression from synchronized EEG-fNIRS data, decomposing affect into a stimulus-shared component and an individual component calibrated via label-free alpha-band cross-channel synchrony, outperforming EEGNet and ASAC-Net baselines while reporting a systematic negative-result search over alternative architectures and features.

Researchers from University of North Texas propose montage and blending fusion techniques for EEG time-frequency images with a lightweight CNN to predict rTMS depression therapy outcomes, and show that near-perfect accuracy under segment-level validation collapses to near-chance under subject-disjoint cross-validation, cautioning about validation methodology in EEG literature.

NeurDuo-EEG is a causal EEG foundation model with channel-resolved persistent state and explicit multi-timescale memory, pre-trained on 3,955 hours of EEG via autoregressive prediction of discrete spectral codes. It achieves top results on four of five benchmarks (including a large seizure-detection AUC-PR jump from 0.285 to 0.471) with a 4.7M-parameter variant supporting efficient streaming inference.

Neurable unveiled the Neurable One, its first standalone brain-scanning headphones featuring 12 EEG sensors that passively monitor brain activity to help users track focus, stress, and other cognitive metrics, priced at $499.

Researchers propose a calibrated hybrid ensemble combining five deep learning and three classical ML models via a logistic regression stacking meta-learner for EEG-based epileptic seizure forecasting. Evaluated on CHB-MIT with Leave-One-Patient-Out cross-validation, the model reaches 74.2% seizure-level sensitivity at 1.24 false alarms per hour with ~16.9 minutes of warning time.

该论文发现非侵入式脑电解码语音中报告的大幅改进很大程度上可由时间间隔捷径在无脑数据的合成信号上复现；通过将句子内的滑动窗口独立编码以移除该捷径后，提出的SimpleB2T方法在五次观测下达到36.6%的词错误率，显著提升了脑到文本解码性能。

This paper presents Signal2Symbol, a neuro-symbolic framework for explainable anomaly detection in physiological time-series data like ECG and EEG, leveraging symbolic tokenization and temporal reasoning to produce interpretable explanations of anomalous patterns.

Bio-MF is a one-step generative framework for EEG-to-fNIRS cross-modal generation, enabling low-latency and high-fidelity synthesis for hybrid motor-imagery brain-computer interfaces.

This paper investigates the effects of cross-depth aggregation and soft-routed experts on EEG foundation model fine-tuning across multiple datasets, showing inconsistent benefits and significant computational overhead.

FRIST is a two-stage EEG decoding framework that leverages fMRI data to improve EEG-only individual-finger BCI decoding, demonstrating increased accuracy in movement execution and motor imagery tasks.