Towards individualized predictions of human sleep and circadian timing

This talk will be presented online. Zoom link: 709 120 4849 (pw: 1234)

Abstract: Accurate assessment of circadian timing is critical to many applications, including timing of drug delivery, prediction of neurobehavioral performance, and optimized scheduling of sleep. Current methods for measuring circadian timing are onerous and do not produce results in real time. Mathematical models have been developed for predicting circadian timing from an individual’s light exposure patterns, which can be applied to passively collected data. These models are now well validated in the field at the group-average level, but tend to perform poorly at the individual level. One potential solution to this problem is the estimation of model parameters at an individual level. We explored whether this approach could be applied to parameters relating to an individual’s light sensitivity. We found that these parameters can account for inter-individual and intra-individual variation in circadian timing. These findings demonstrate that model parametrization based on physiological measurements of light sensitivity could lead to more accurate individual-level circadian phase prediction.