CaMBRAIN: Real-time, Continuous EEG Inference with Causal State Space Models

2026-05-27

AI Summary

This work addresses the challenge of efficiently modeling long-duration, continuous EEG signals for real-time inference, a task hindered by the quadratic complexity of attention mechanisms and limitations of sliding-window strategies in existing deep learning approaches. To overcome this, we introduce the causal Mamba state space model into EEG analysis for the first time and propose a multi-stage self-supervised training framework tailored for streaming processing. Our method achieves linear time complexity while enabling long-range, continuous modeling of EEG sequences of arbitrary length. Evaluated on three standard EEG benchmarks, the approach attains state-of-the-art performance and demonstrates over a 10× improvement in inference throughput, thereby enabling, for the first time, real-time, causal, and long-range inference on variable-length EEG signals.

Abstract

Electroencephalography (EEG) is a critical, non-invasive method to monitor electrical brain activity. EEGs can span anywhere from a couple seconds to multiple hours, posing a major hurdle for existing deep learning methods due to two major factors: (1) existing EEG models are predominantly built upon the attention mechanism, incurring quadratic scaling as the sequence length increases, and (2) raw EEG signals must be processed in a sliding-window fashion due to fixed-length input requirements, preventing global understanding of the entire signal. To this extent, we propose CaMBRAIN - the first Causal, Mamba-based state space model (SSM) capable of real-time inference of EEG signals, arguing that bidirectional approaches are needlessly expensive given the causal, unidirectional nature of EEG. However, training such a model is non-trivial, as crucial EEG events can be extremely brief - within fractions of a second - yet separated by long intervals spanning minutes. Current EEG methods use self-supervised objectives that optimize for signal reconstruction, but these are not well suited for streaming SSMs; they fail to explicitly train the hidden state to retain the salient long-range context needed for streaming inference. We therefore introduce a multi-stage self-supervised training pipeline specifically tailored to encourage long-range memory retention and strong performance on EEG signals, while preserving the linear-time complexity of state space models. CaMBRAIN achieves state-of-the-art (SOTA) results across 3 different EEG datasets with >10x higher throughput than existing models, enabling the first model capable of long-range, continuous inference of variable-length EEG signals.

Problem

• EEG
• real-time inference
• long-range dependency
• continuous signal
• scalability

Innovation

• Causal State Space Models
• Real-time EEG Inference
• Mamba Architecture
• Long-range Memory
• Self-supervised Training