Re-evaluating two popular EEG-based mobile sleep-monitoring devices for home use

Area of Science:

• Consumer sleep technology
• Sleep science
• Biomedical engineering

Background:

• Mobile sleep-monitoring devices offer accessible sleep analysis.
• Traditional polysomnography requires specialized labs and technicians.
• Independent accuracy evaluations of consumer sleep devices are limited.

Purpose of the Study:

• To compare the performance of two popular electroencephalogram-based mobile sleep systems: DREEM 3 and Zmachine Insight+.
• To assess the accuracy of these devices against each other and sleep logs.
• To establish a Bayesian lower limit for expected sleep stage agreement.

Main Methods:

• 25 participants simultaneously used DREEM 3 and Zmachine Insight+ for two nights.
• Sleep logs were maintained by participants.
• Device performance was compared using standard sleep metrics.
• Bayesian analysis was used to determine expected sleep stage agreement.

Main Results:

• Zmachine Insight+ tended to overestimate wakefulness, potentially misclassifying N1/N2 sleep stages.
• DREEM 3 tended to underestimate wakefulness, mistaking it for N1/N2 sleep stages.
• Device agreement improved from the first to the second night of use.

Conclusions:

• Both DREEM 3 and Zmachine Insight+ exhibit discrepancies in sleep stage scoring compared to previous developer claims.
• User training and device usage protocols may influence accuracy.
• Recommendations are provided for optimizing the use of these consumer sleep-monitoring devices.