BrainFuseNet: Enhancing Wearable Seizure Detection Through EEG-PPG-Accelerometer Sensor Fusion and Efficient Edge Deployment

Abstract

BrainFuseNet enhances wearable seizure detection by fusing EEG, PPG, and accelerometer signals, combining complementary physiological and motion information to reduce artifact-induced false alarms while maintaining high seizure sensitivity. The system is co-designed for efficient edge deployment on ultra-low-power processors, validated on clinical data, and demonstrates real-time operation within the power budget of long-term wearable epilepsy monitoring.

Thorir Mar Ingolfsson

Postdoctoral Researcher

I develop efficient machine learning systems for biomedical wearables that operate under extreme resource constraints. My work bridges foundation models, neural architecture design, and edge deployment to enable real-time biosignal analysis on microwatt-scale devices.