Slowed eyelid closure coupled with increased duration and frequency of closure is associated with drowsiness. This study assessed the utility of two devices for automated measurement of slow eyelid closure in a standard poor performance condition (alcohol) and following 12-h sleep deprivation. Twenty-two healthy participants (mean age=20.8 (SD 1.9) years) with no history of sleep disorders participated in the study. Participants underwent one baseline and one counterbalanced session each over two weeks; one 24-hour period of sleep deprivation, and one daytime session during which alcohol was consumed after a normal night of sleep. Participants completed a test battery consisting of a 30-min simulated driving task, a 10-min Psychomotor Vigilance Task (PVT) and the Karolinska Sleepiness Scale (KSS) each in two baseline sessions, and in two randomised, counterbalanced experimental sessions; following sleep deprivation and following alcohol consumption. Eyelid closure was measured during both tasks using two automated devices (Copilot and Optalert™). There was an increase in the proportion of time with eyelids closed and the Johns Drowsiness Score (incorporating relative velocity of eyelid movements) following sleep deprivation using Optalert (p<0.05 for both). These measures correlated significantly with crashes, PVT lapses and subjective sleepiness (r-values 0.46-0.69, p<0.05). No difference between the two sessions for PERCLOS recorded during the PVT or the driving task as measured by the Copilot. The duration of eyelid closure predicted frequent lapses following sleep deprivation (which were equivalent to the average lapses at a blood alcohol concentration of 0.05% - area under curve for ROC curve 0.87, p<0.01). The duration of time with slow eyelid closure, assessed by the automated devices, increased following sleep deprivation and was associated with deterioration in psychomotor performance and subjective sleepiness. Comprehensive algorithms inclusive of ocular parameters may be a better indicator of performance impairment following sleep loss.
The utility of automated measures of ocular metrics for detecting driver drowsiness during extended wakefulness
2016
Article (Journal)
English
A FPGA based driver drowsiness detecting system
IEEE | 2005
|Machine Learning Systems for Detecting Driver Drowsiness
British Library Conference Proceedings | 2009
|Machine Learning Systems for Detecting Driver Drowsiness
Springer Verlag | 2008
|