Related Experiment Video
Updated: May 7, 2025
09:58
Recording Brain Activity with Ear-Electroencephalography
Published on: March 31, 2023
2.7K
Comparison analysis between standard polysomnographic data and in-ear-electroencephalography signals: a preliminary
Gianpaolo Palo1,2, Luigi Fiorillo1, Giuliana Monachino1,3
1Department of Innovative Technologies, Institute of Digital Technologies for Personalized Healthcare (MeDiTech), University of Applied Sciences and Arts of Southern Switzerland, Lugano, Switzerland.
Sleep Advances : a Journal of the Sleep Research Society
|December 30, 2024
Summary
In-ear electroencephalography (EEG) offers a promising, less invasive alternative for sleep disorder monitoring. While hypnogram agreement showed variability, in-ear EEG signals demonstrated high similarity to polysomnography (PSG) in feature analysis.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Sleep Medicine
Background:
- Polysomnography (PSG) is the gold standard for sleep disorder evaluation but is uncomfortable for long-term use, introducing bias.
- The need for less invasive, portable, and cost-effective sleep monitoring alternatives is critical.
- In-ear electroencephalography (EEG) sensors present a potential solution for unobtrusive sleep tracking.
Purpose of the Study:
- To develop and validate a methodology for assessing the similarity between single-channel in-ear EEG and standard PSG derivations.
- To compare hypnogram agreement and signal features between in-ear EEG and PSG recordings.
Main Methods:
- Collected 4-hour sleep recordings from 10 healthy adults (aged 18-60).
- Analyzed data using hypnogram-based agreement (Cohen's kappa, Fleiss' kappa) and feature-based similarity (Jensen-Shannon Divergence Feature-based Similarity Index - JSD-FSI).
- Extracted time- and frequency-domain features and performed unsupervised feature selection.
Main Results:
- Hypnogram agreement showed significant variability between PSG and in-ear EEG scorers.
- Despite hypnogram variability, in-ear EEG signals exhibited high JSD-FSI similarity to PSG across sleep stages (awake: 0.79±0.06, NREM: 0.77±0.07, REM: 0.67±0.10).
- Signal similarity was comparable to that found between standard PSG channel combinations.
Conclusions:
- In-ear EEG is a viable technology for home-based sleep monitoring.
- Further research with larger, more diverse datasets is required to fully establish its clinical utility.
- In-ear EEG shows potential to overcome limitations of traditional PSG for sleep quality assessment.
