Early detection of falling asleep at the wheel: A Heart Rate Variability approach

In this study we check the feasibility of a new ECG-based approach to detect driverspsila propensity to fall asleep at the wheel. Ten healthy volunteers, under conditions of increasing sleep deprivation (up to 34 hours), were asked to alternately undergo a Maintenance of Wakefulness Test or a Driving Simulation test every 2 hours while ECG, EEG, EMG, eye movement and video were recorded. Results from 59 falling asleep (FA) events tracked from the first 5 volunteers during MWT provide promising trends: Heart Rate Variability in the VLF range decreases consistently and significantly minutes before FA events. The sympatho-vagal balance is very low compared to baseline wake values for about 5 minutes before the events. The mean HR and overall RR variability decrease during FA events by 2.2 SD and 2.9 SD below regional means. These changes found during MWT suggest that ECG derived parameters in the time and time-frequency domains may provide a useful tool for monitoring driverspsila drowsiness and preventing traffic accidents.