Empirical evaluation of consumer EEG and actigraphy devices for home-based sleep assessment

Abstract

Sleep is the most essential part of everyone's daily life. With prevalence of wearable and mobile technologies, several consumer solutions have emerged, targeting home-based sleep assessment. This study aims to investigate the performance of EEG and actigraphy devices in assessing sleep quality. Various ambient factors affecting sleep quality and usability of such wearable devices were also evaluated. The study protocol consists of an online opinion survey, subjective sleep assessment and multi-night data recording with or without white noise. From survey results, 94.2% of respondents had no prior experience in using sleep-tracking devices but 37.3% of total respondents encounter sleep problems. We collected multi-night sleep data with 18 high school student subjects. Significant correlation was found between sleep parameters and factors like naps, caffeine and stress. Playing white noise during sleep showed improvement in the occurrence and duration of deep sleep, possibly a positive effect of sound-based sleep assistance. Our analysis showed the sleep quality parameters derived from EEG are more complete and accurate than actigraphy. However, actigraphy surpassed EEG headband in usability aspects such as comfort, ease of use. Despite that, outcomes from our product design survey showed no significance difference in preference between eye-mask and wristband. We hope that our findings contribute to further development of home-based sleep solution with better usability, reliable sleep assessment, for early identification and treatment of sleep related problems.