发表机构
Berlin Institute for the Foundations of Learning and Data; Technische Universität Berlin; Korea University; Max Planck Institut für Informatik; Basque Center on Cognition, Brain and Language; Ikerbasque（柏林学习与数据基础研究所; 柏林工业大学; 高丽大学; 马克斯·普朗克信息学研究所; 巴斯克认知、大脑与语言中心; 伊克尔巴斯克基金会）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对EEG源成像的时空权衡，提出双流框架解耦全局时间编码与逐点空间细化，以时间先验提升定位精度，并验证了跨几何迁移及真实数据上的年龄解码能力。
AI 中文摘要
脑电图（EEG）提供毫秒级的时间分辨率，但推断潜在的神经源是一个严重不适定的空间逆问题。虽然深度学习已推进空间重建，但当前架构面临一个关键困境：逐帧模型丢弃了重要的时间上下文，而完整的4D时空网络则在重建精度和推理成本之间引入了架构权衡。我们提出了一种新颖的双流框架，将全局时间表示学习与逐时间点的空间细化明确解耦。基于Transformer的时间条件编码器通过因子化的时空注意力处理整个EEG序列，保留传感器分辨的特征。一个固定的逆算子随后将这些特征映射为源索引的条件，用于逐时间步的源空间Transformer或体积卷积细化器。在逼真合成数据上的广泛评估表明，这种时间先验显著改善了空间定位，优于经典和时空基线，特别是在高噪声和多源场景中。跨多种导联场和显式算子不匹配的训练提高了对未见头部几何的迁移能力，并使基于模板的重建更接近个体化反演。此外，我们将仅在合成EEG数据上训练的模型应用于真实世界EEG。在睁眼、闭眼条件下，基于源功率差异拟合的逻辑回归器成功解码了年龄组。
英文摘要
Electroencephalography (EEG) offers millisecond temporal resolution, but inferring underlying neural sources is a severely ill-posed spatial inverse problem. While deep learning has advanced spatial reconstruction, current architectures face a critical dilemma: frame-by-frame models discard vital temporal context, whereas full 4D spatiotemporal networks introduce an architectural trade-off between reconstruction accuracy and inference cost. We propose a novel two-stream framework that explicitly decouples global temporal representation learning from per-time-point spatial refinement. A Transformer-based Temporal Condition Encoder processes the entire EEG sequence via factorized spatiotemporal attention, retaining sensor-resolved features. A fixed inverse then maps these features into source-indexed conditioning for a per-timestep Source-Space Transformer or volumetric convolutional refiner. Extensive evaluations on realistic synthetic data demonstrate that this temporal prior dramatically improves spatial localization, outperforming classical and spatiotemporal baselines, particularly in high-noise and multi-source regimes. Training across diverse leadfields and explicit operator mismatches improves transfer to unseen head geometries and brings template-based reconstruction closer to subject-specific inversion. Furthermore, we apply the model trained only on synthetic EEG data to real-world EEG. A logistic regressor fit on source power differences in eyes-open, eyes-closed conditions successfully decodes age groups.
CommentsThis work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
