Nonlinear Dynamical Modeling of Human Intracranial Brain Activity with Flexible Inference

2025-12-28

AI Summary

Linear models struggle to capture nonlinear dynamics in intracranial electroencephalography (iEEG) multichannel signals, while mainstream RNNs lack robustness to missing data during inference. Method: This work extends the DFINE framework to iEEG modeling for the first time, proposing a hybrid deep state-space model that integrates nonlinear neural networks with linear state-space structure. It employs joint optimization, band-specific analysis—focusing on the high-gamma band—and a robust inference mechanism for missing observations. Results: The model significantly outperforms conventional linear state-space models on iEEG prediction tasks, matches or exceeds GRU performance, and demonstrates superior robustness to both random and structured missing data. Notably, modeling accuracy in the high-gamma band is substantially improved. This advances real-time, robust next-generation brain–computer interfaces by providing an interpretable, deployable dynamical modeling paradigm.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: State Estimation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Dynamical modeling of multisite human intracranial neural recordings is essential for developing neurotechnologies such as brain-computer interfaces (BCIs). Linear dynamical models are widely used for this purpose due to their interpretability and their suitability for BCIs. In particular, these models enable flexible real-time inference, even in the presence of missing neural samples, which often occur in wireless BCIs. However, neural activity can exhibit nonlinear structure that is not captured by linear models. Furthermore, while recurrent neural network models can capture nonlinearity, their inference does not directly address handling missing observations. To address this gap, recent work introduced DFINE, a deep learning framework that integrates neural networks with linear state-space models to capture nonlinearities while enabling flexible inference. However, DFINE was developed for intracortical recordings that measure localized neuronal populations. Here we extend DFINE to modeling of multisite human intracranial electroencephalography (iEEG) recordings. We find that DFINE significantly outperforms linear state-space models (LSSMs) in forecasting future neural activity. Furthermore, DFINE matches or exceeds the accuracy of a gated recurrent unit (GRU) model in neural forecasting, indicating that a linear dynamical backbone, when paired and jointly trained with nonlinear neural networks, can effectively describe the dynamics of iEEG signals while also enabling flexible inference. Additionally, DFINE handles missing observations more robustly than the baselines, demonstrating its flexible inference and utility for BCIs. Finally, DFINE's advantage over LSSM is more pronounced in high gamma spectral bands. Taken together, these findings highlight DFINE as a strong and flexible framework for modeling human iEEG dynamics, with potential applications in next-generation BCIs.

Problem

• Modeling nonlinear dynamics in intracranial EEG for brain-computer interfaces
• Enabling flexible inference with missing neural data in recordings
• Extending deep learning frameworks to multisite human iEEG signals

Innovation

• Extends DFINE framework to intracranial EEG modeling
• Integrates neural networks with linear state-space models
• Enables flexible inference with robust missing data handling