REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates
This addresses the challenge of real-time implementation in clinical devices for epilepsy patients, though it appears incremental as it builds on existing graph neural network and recurrent structure approaches.
The paper tackles the problem of slow inference speed and high memory usage in EEG-based seizure detection models by introducing REST, a graph-based residual state update mechanism. The result is a 9-fold acceleration in inference speed compared to state-of-the-art models while using substantially less memory.
EEG-based seizure detection models face challenges in terms of inference speed and memory efficiency, limiting their real-time implementation in clinical devices. This paper introduces a novel graph-based residual state update mechanism (REST) for real-time EEG signal analysis in applications such as epileptic seizure detection. By leveraging a combination of graph neural networks and recurrent structures, REST efficiently captures both non-Euclidean geometry and temporal dependencies within EEG data. Our model demonstrates high accuracy in both seizure detection and classification tasks. Notably, REST achieves a remarkable 9-fold acceleration in inference speed compared to state-of-the-art models, while simultaneously demanding substantially less memory than the smallest model employed for this task. These attributes position REST as a promising candidate for real-time implementation in clinical devices, such as Responsive Neurostimulation or seizure alert systems.