To ensure mission success, astronauts must maintain a high level of performance, even when work-rest schedules result in chronic sleep restriction and circadian misalignment, both of which can contribute to fatigue and performance deficits unless effective countermeasures are used. The overarching goal of this project is to build an Individualized Fatigue Meter to provide astronauts and mission support personnel with immediate feedback about their alertness or fatigue levels. The Individualized Fatigue Meter incorporates sleep history, ambient light levels, schedule information, and performance tests to provide immediate individualized feedback about alertness, and allow critical insight when making decisions about scheduling future sleep periods and selecting fatigue countermeasures.
The Individualized Fatigue Meter software prototype is designed to inform astronauts and mission support personnel about current and future alertness or fatigue levels. The supporting algorithms necessary to process and present this fatigue information include a model-independent computational architecture, a hybrid biomathematical fatigue model that is specifically designed for space exploration missions, and a data fusion algorithm that statistically combines multiple sources of sleep data in an optimal way.
The Individualized Fatigue Meter meets the specific requirements of long-duration exploration missions, and provides feedback to astronauts and flight surgeons about fatigue, as well as aids in the selection of fatigue countermeasures. When validated and deployed as part of the Behavioral Health and Performance Dashboard Software, the Individualized Fatigue Meter will support crew behavioral health during training and missions.
Systems and methods were developed to estimate a subject’s actual sleep status over time by applying data fusion algorithms to sleep data sets collected from multiple sleep data sources. Embodiments employ various algorithms incorporating Bayes Theorem to combine sleep data from actigraphy, sleep diary, direct observation, sleep schedules, work schedules, performance tests, neurobehavioral tests, and the like. Particular embodiments assign data error characteristics to each source, determine likelihoods of correct reporting of sleep status from each source, and apply Bayesian analysis to each source-specific likelihood to determine an overall sleep status estimate.
This work was done by Kevin Kan, Christopher Mott, Daniel Mollicone, and Michael Stubna of Pulsar Informatics, Inc. for Johnson Space Center. MSC-25622-1