TGSD: Topology-Guided State-Space Diffusion for EEG Spatial Super-Resolution

2026-05-22

AI Summary

This work addresses the challenge of insufficient spatial information in low-density EEG due to sparse electrode placement, which hinders accurate characterization of cross-regional neural activity. To overcome this limitation, the authors propose the Topology-Guided Spatial Diffusion (TGSD) framework, which integrates a hierarchical spatial prior encoder that fuses local geometric structure with regional contextual information to construct a full-electrode topological prior. Furthermore, they introduce a conditional state-space diffusion reconstructor that alternately models long-range dependencies along temporal and channel dimensions during the reverse diffusion process, enabling high-fidelity spatial super-resolution reconstruction. TGSD is the first to combine topology-aware priors with conditional diffusion mechanisms and leverages state-space models to jointly capture EEG’s temporal dynamics and inter-channel dependencies, effectively mitigating reconstruction ambiguity under missing channels. Experiments on SEED and PhysioNet MM/I datasets demonstrate that TGSD consistently outperforms existing methods across multiple super-resolution scales, enhancing both reconstruction quality and downstream classification performance, thereby validating its practical utility in wearable EEG applications.

Abstract

Low-density EEG is more suitable for wearable and IoT-based brain sensing, but sparse electrode sampling often lacks sufficient spatial information to characterize cross-regional neural activity. EEG spatial super-resolution aims to recover dense-channel EEG from sparse recordings, yet remains challenging because channel missingness typically occurs at the whole-channel level, spatiotemporal dependencies over the full electrode layout are often underexplored, and the mapping from sparse to dense signals is inherently ambiguous. To address these issues, we propose TGSD, a topology-guided state-space diffusion framework for EEG spatial super-resolution. TGSD first employs a Hierarchical Spatial Prior Encoder to learn topology-aware priors over the complete electrode layout by integrating local geometric relationships with region-level contextual information. Based on these priors and sparse observations, a Conditional State-Space Diffusion Reconstructor progressively generates missing-channel signals through reverse diffusion, while alternating temporal and channel-wise state-space modeling captures long-range temporal dynamics and inter-channel dependencies in a unified framework. Experiments on the SEED and PhysioNet MM/I datasets show that TGSD consistently outperforms representative baselines under different super-resolution factors in both reconstruction fidelity and downstream classification performance. These results demonstrate the effectiveness of combining topology-aware spatial priors with conditional diffusion for enhancing practical low-density EEG sensing in wearable and IoT scenarios. The official implementation code is available at https://github.com/jtggz/TGSD.

Problem

• EEG spatial super-resolution
• low-density EEG
• sparse electrode sampling
• spatiotemporal dependencies
• channel missingness

Innovation

• topology-aware prior
• state-space diffusion
• EEG spatial super-resolution
• hierarchical spatial encoding
• conditional generation