We can’t show the full text here under this license. Use the link below to read it at the source.
Prediction of shiftworker alertness, sleep, and circadian phase using a model of arousal dynamics constrained by shift schedules and light exposure
Predicting shift workers’ alertness, sleep, and body clock timing from work schedules and light exposure
AI simplified
Abstract
Combinations including light information predicted urinary 6-sulphatoxymelatonin acrophase within ±1 hour for 65% of nurses on diurnal schedules.
- Simulations showed that 56%-60% of subjective sleepiness scores were predicted within ±1 on a day shift and 48%-52% on a night shift.
- Accurate prediction of circadian phase required individualized light input.
- Minute-by-minute sleep-wake state overlap between the model and empirical data ranged from 81 ± 6% to 87 ± 5%, depending on input constraints.
- The model's predictions were more accurate when including environmental factors such as light exposure.
AI simplified
Key numbers
56%
KSS Prediction Accuracy Day Shift
Percentage of KSS scores predicted within ±1 on the day shift.
48%
KSS Prediction Accuracy Night Shift
Percentage of KSS scores predicted within ±1 on the night shift.
65%
Circadian Phase Prediction Accuracy
Percentage of aMT6s acrophase values predicted within ±1 h.