Back to Home

Machine Learning for Cognitive and Auditory Response Analysis Using Sleep and Wearable EEG

EEG Signal Processing Feature Extraction Spectral Analysis Machine Learning Statistical Modeling Wearable Sensors Biomarker Development

Project Overview

This project combines standard overnight EEG with in-ear wearable EEG recordings to investigate how sleep physiology reflects cognitive health and how auditory stimulation influences slow-wave activity. The work includes extracting spectral and temporal biomarkers from continuous sleep EEG, modeling slow-wave responses to sound cues, and applying classification methods to detect signatures of cognitive impairment. Through machine learning, signal processing, and clinical interpretation, the project demonstrates how sleep-based neural activity—captured from both in-lab and wearable sensors—can serve as a sensitive indicator of neurocognitive status and support next-generation digital health and brain-monitoring technologies.

Related Publications

Average EEG waveforms showing enhanced slow-wave activity during auditory stimulation compared to sham

Average EEG waveforms showing enhanced slow-wave activity during auditory stimulation compared to sham. View Interactive Demo

Wearable in-ear EEG device used for continuous sleep monitoring

Wearable in-ear EEG device used for continuous sleep monitoring.