ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks

TL;DR

ODEBRAIN introduces a continuous-time neural ODE framework for modeling EEG brain networks, capturing instantaneous nonlinear dynamics and improving forecasting accuracy over traditional discretized methods.

Contribution

It presents a novel Neural ODE-based approach that integrates spectral graph features for continuous and robust modeling of brain network dynamics from EEG data.

Findings

Significantly improves EEG dynamic forecasting accuracy

Enhances robustness and generalization in brain network modeling

Captures instantaneous nonlinear brain activity variations

Abstract

Modeling neural population dynamics is crucial for foundational neuroscientific research and various clinical applications. Conventional latent variable methods typically model continuous brain dynamics through discretizing time with recurrent architecture, which necessarily results in compounded cumulative prediction errors and failure of capturing instantaneous, nonlinear characteristics of EEGs. We propose ODEBRAIN, a Neural ODE latent dynamic forecasting framework to overcome these challenges by integrating spatio-temporal-frequency features into spectral graph nodes, followed by a Neural ODE modeling the continuous latent dynamics. Our design ensures that latent representations can capture stochastic variations of complex brain states at any given time point. Extensive experiments verify that ODEBRAIN can improve significantly over existing methods in forecasting EEG dynamics with enhanced robustness and generalization capabilities.