A real-time, personalized sleep intervention using mathematical modeling and wearable devices.

Area of Science:

• Chronobiology
• Sleep Science
• Human-Computer Interaction

Background:

• Increased artificial light exposure necessitates extended periods of alertness beyond traditional daytime hours.
• Shift work and complex schedules disrupt natural sleep-wake patterns, impacting performance and health.
• Wearable technology offers opportunities for real-world monitoring of sleep-wake dynamics.

Purpose of the Study:

• To develop a personalized sleep intervention framework using wearable device data.
• To maximize user alertness during specific target periods, especially for shift workers.
• To discover and validate novel sleep-wake patterns.

Main Methods:

• Analysis of real-world sleep-wake patterns from wearable devices.
• Development of a mathematical model to track dynamic sleep pressure and circadian rhythm.
• Validation of the framework with shift workers (N=71, t=13-21 days).
• Creation of a mobile application for personalized sleep schedule recommendations.

Main Results:

• Accurate prediction of real-time alertness, even for complex shift schedules.
• Discovery of the 'adaptive circadian split sleep' pattern, combining a main sleep period with a late nap.
• Demonstrated ability to enhance alertness during both work and non-work periods for shift workers.

Conclusions:

• The personalized sleep intervention framework effectively maximizes alertness for individuals with demanding schedules.
• The adaptive circadian split sleep pattern offers a novel approach to managing sleep for improved shift worker performance.
• The integrated mobile application provides practical, personalized sleep recommendations to enhance health and reduce errors.