Reducing the Complexity of Deep Learning Models for EEG Analysis on Wearable Devices

2026-06-10

AI Summary

This study addresses the challenge of deploying complex deep learning models for real-time EEG-based seizure detection on wearable devices, which are constrained by limited computational power, energy budget, and memory bandwidth. The authors systematically investigate the trade-off between model accuracy and complexity, proposing a lightweighting strategy that combines parameter quantization with a reduction in the number of EEG electrodes. This approach is rigorously evaluated across multiple state-of-the-art EEG-specific deep neural networks. Experimental results demonstrate that the proposed method substantially reduces computational overhead and resource requirements while incurring only minimal accuracy degradation, thereby offering a practical and efficient deployment solution for real-time EEG analysis in wearable settings.

Abstract

Wearable healthcare devices are the fastest-growing Internet of Things (IoT) sector. Many automated healthcare services rely on two crucial biological signals, namely ECG and EEG, which reflect the activity of the heart and brain, respectively. Although deep neural networks are considered the primary way to process and analyze these signals, the very tight energy and computational power constraints in wearable devices are far below the computational, energy, and memory bandwidth demands of DNN models, thereby impeding the deployment of deep learning in many practical wearable services. This paper investigates the feasibility of deploying state-of-the-art DNN models in resource-constrained wearable devices. Notably, we explore the trade-off between accuracy and computational complexity of DNNs when parameter quantization and electrode reduction methods are used. Our investigation centers on several state-of-the-art DNN models designed for EEG signal analysis, specifically for detecting epileptic seizures. Our findings demonstrate that, when applied judiciously, these techniques can significantly reduce the complexity of the DNNs under consideration with minimal adverse effects on accuracy. These results reveal the explicit trade-offs between accuracy and complexity reduction encountered when adapting DNN-based online EEG analysis for wearable devices.

Problem

• EEG analysis
• wearable devices
• deep learning
• computational complexity
• resource constraints

Innovation

• parameter quantization
• electrode reduction
• EEG analysis
• wearable devices
• model complexity reduction