Comparison of sleep-wake classification using electroencephalogram and wrist-worn multi-modal sensor data

Annu Int Conf IEEE Eng Med Biol Soc. 2014:2014:930-3. doi: 10.1109/EMBC.2014.6943744.

Abstract

This paper presents the comparison of sleep-wake classification using electroencephalogram (EEG) and multi-modal data from a wrist wearable sensor. We collected physiological data while participants were in bed: EEG, skin conductance (SC), skin temperature (ST), and acceleration (ACC) data, from 15 college students, computed the features and compared the intra-/inter-subject classification results. As results, EEG features showed 83% while features from a wrist wearable sensor showed 74% and the combination of ACC and ST played more important roles in sleep/wake classification.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accelerometry
  • Electroencephalography / instrumentation*
  • Electroencephalography / methods*
  • Galvanic Skin Response / physiology
  • Humans
  • Polysomnography
  • ROC Curve
  • Skin Temperature / physiology
  • Sleep / physiology*
  • Wakefulness / physiology*
  • Wrist / physiology*
  • Young Adult