ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks
#ODEBrain #Neural ODE #EEG modeling #brain networks #dynamic forecasting #spatio-temporal features #neuroscience research #clinical applications
📌 Key Takeaways
- Researchers developed ODEBRAIN, a Neural ODE framework for brain network modeling
- Traditional methods fail to capture EEG's instantaneous, nonlinear characteristics
- ODEBRAIN integrates spatio-temporal-frequency features with Neural ODE modeling
- The framework shows significant improvements in EEG forecasting with enhanced robustness
📖 Full Retelling
Researchers Haohui Jia and eight collaborators have developed ODEBRAIN, a novel Neural ODE framework for modeling dynamic brain networks, as announced in their paper published on arXiv on February 26, 2026. The research addresses critical limitations in conventional methods used to model neural population dynamics, which are essential for both foundational neuroscience research and various clinical applications. Traditional latent variable approaches typically discretize time using recurrent architectures, resulting in compounded cumulative prediction errors and an inability to capture the instantaneous, nonlinear characteristics of EEG signals. The ODEBRAIN framework overcomes these challenges by integrating spatio-temporal-frequency features into spectral graph nodes, followed by a Neural ODE component that models continuous latent dynamics. This innovative design ensures that the latent representations can effectively capture stochastic variations of complex brain states at any given time point. The researchers conducted extensive experiments that demonstrate ODEBRAIN's significant improvements over existing methods in forecasting EEG dynamics, with enhanced robustness and generalization capabilities that could revolutionize both research and clinical applications in neuroscience.
🏷️ Themes
Neuroscience, Artificial Intelligence, Medical Technology
📚 Related People & Topics
Neural differential equation
Equation in machine learning
Neural differential equations are a class of models in machine learning that combine neural networks with the mathematical framework of differential equations. These models provide an alternative approach to neural network design, particularly for systems that evolve over time or through continuous ...
Entity Intersection Graph
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2602.23285 [Submitted on 26 Feb 2026] Title: ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks Authors: Haohui Jia , Zheng Chen , Lingwei Zhu , Rikuto Kotoge , Jathurshan Pradeepkumar , Yasuko Matsubara , Jimeng Sun , Yasushi Sakurai , Takashi Matsubara View a PDF of the paper titled ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks, by Haohui Jia and 8 other authors View PDF HTML 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. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2602.23285 [cs.AI] (or arXiv:2602.23285v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2602.23285 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Haohui Jia [ view email ] [v1] Thu, 26 Feb 2026 17:59:10 UTC (29,557 KB) Full-text links: Access Paper: View a PDF of the paper titled ODEBrain: Continuous-Time EEG Graph for Modeling Dynamic Brain Networks, by Haohui Jia and 8 other authors View PDF HTML TeX Source view license Current brow...
Read full article at source