A geometry aware framework enhances noninvasive mapping of whole human brain dynamics

A team of researchers led by Song Wang has introduced a geometry-aware framework that significantly enhances noninvasive mapping of whole-brain dynamics using electroencephalography (EEG) and magnetoencephalography (MEG). The method, detailed in a preprint on arXiv, embeds participant-specific Geometric Basis Functions (GBFs)—eigenmodes derived from each individual's cortical surface—to provide a powerful anatomic constraint. This resolves the long-standing inverse problem in source imaging, where simplistic or biologically implausible priors have limited reconstruction fidelity. By representing neural sources as linear combinations of geometric basis functions, GBF aligns source estimates with the geometric organization of neural dynamics.

Validation across the Meta-Source Benchmark, task-evoked data, resting-state networks, intracranial stimulation, and epilepsy data demonstrates that GBF yields high localization accuracy and captures fast spatiotemporal dynamics consistent with anatomical pathways. The findings suggest that both spontaneous and evoked whole-brain activity can be described by hundreds of geometric modes, offering a compact yet accurate representation of neural sources. This approach links cortical geometry to electrophysiological dynamics, providing a versatile source imaging tool for both scientific and clinical applications, potentially revolutionizing noninvasive brain mapping.