Sleep Shift Scheduling Tool

Sleep loss and fatigue may lead to reduced performance and an increased risk to safety during many activities, including spaceflight. Your challenge is to develop an operational sleep shift scheduling tool that provides autonomous customization of a schedule for sleep, exercise, and nutrition to manage fatigue.

Sleep Better

Summary

Our app preserves the integrity of the users' sleep cycles. Using non intrusive sensors that detect when the astronaut start sleeping, our application will then decide when to wake him or her up, therefore removing the fatigue induced by uneven sleep cycles.

How We Addressed This Challenge

We developed a solution to provide the optimal user sleep schedule.

The aim of this project is to design the optimal schedule needed by the user to be in shape and efficient. This uses the user scores for cognitive and physical daily tests.


We based our solution on an AI that analyzes the users' sleep routines.

With deep learning, the AI will plan the optimal amount of sleep cycles per night in regards to the efficiency of the users own number of sleep cycles. For example, if we detect that the user should sleep 5 cycles instead of his usual 3 cycles, then we will organize his/her routine to acclimatize her/him to sleep that amount of cycles.


AI may also adapt someone's sleep schedule for an incoming jetlag, by decreasing the number of sleep cycles aloted in a row and increasing the number of naps, therefore tailoring a custom polyphasic sleep schedule to the user. This pattern is based on already existing polyphasic methods.


Data necessary for the proejct could be gathered using any IOT devices that can track users medical details with devices such as smartwatches, or the Dreem 2 (https://dreem.com) among others.

How We Developed This Project

We have been inspired by our personal experiments. We tested polyphasic sleep, long time period sleeping, short time period sleeping, etc... We noticed all effects these experimentation had on us, and we found a major problem: the difficulty to adopt an efficient sleep method when it's different from our usual routines, like when one falls asleep 10 minutes into a 15 minutes nap (resulting in only a 5 minute sleep time overall), or when one has to check something important before going to bed, shifting his/her whole schedule and disrupting the whole organization for the foreseeable day to come.


The idea was a simple tool with complex functionalities.


The user only has to link his everyday IOT devices.

We need only a night to propose a new schedule. Indeed, based on the measurements taken by the IOT devices during that night, the app can obtain the length of REM sleep the user has had with each cycle. If it is found that during the last cycles the length of REM sleep has decreased, the application will propose to replace those last cycles with a nap during the day. This nap proposal will also be made in accordance with any day planning that the user has inputted (for example :google agenda).


With the app or the website the user could launch physical and cognitive tests. He would just need to link his IOT devices such as smartwatches or other IOT devices like Dreem 2 (neurologic IOT device). With these piece of hardware we can get the data we need to provide interesting results for the schedule.

How We Used Space Agency Data in This Project

We understood one of the important aspects of this project by reading several studies on astronauts and the sleeping process, and the well-being that comes with a restful sleep.


First of all, we focused on the activities' organization. As said in the article "Seven Ways Astronauts Improve Sleep May Help You Snooze Better on Earth",

"Properly-timed exercise, minimizing light from digital devices in the evening, and thoughtful dietary choices can all lead to a better night’s sleep and help to prevent circadian misalignment."

Our goal was to organize the activities thanks to several parameters, like 

- the priority of the activity

- the number of person needed to do this activity

- the number of person that can sleep simultaneously

- the rhythm of every member of the group.


The direction we took at the beginning of the project had led us to several questions. One of the important one was, how are we going to understand the rhythm of the users ? One study we’ve read over and over has drawn our attention : "The relation of Eye Movements During Sleep to dream activity : an objective method for the study of dreaming" written by William Dement and Nathaniel Kleitman.

We discovered the different steps of the sleep cycle and its relation with the eye movements, and later on, its link with the cerebral activity with experiments on EEG : "Regularly Occurring Periods of Eye Motility, and Concomitant Phenomena,During Sleep" written by Eugene Aserinsky and Nathaniel Kleitman.

So apart from the very used EEG, we searched for different tools and technologies we could use as a smaller tool of detection of the sleep cycle.

As you may know, Apple with its Apple Watch has already filed a patent for the detection of myoelectric signals, there are already really efficient eye movement trackers, and new solutions as the ones developed by the society "dreem" which would provide a small connected object to help you sleep.


How we are going to protect the well-being of the astronauts was another question which took us some time.

As said in the "NASA Research Reveals Biological Clock Misalignment Effects on Sleep for Astronauts", conserving their biological clocks is an important factor of their sleep cycle.

"When crewmembers slept when their biological clocks were promoting sleep, they slept longer and used fewer sleep medications. When they slept when their biological clocks were not promoting sleep, they had more disrupted sleep and used more sleep medication."

So we first thought having several tests before going in orbit could let our AI predict their sleep habits, and adjust their schedule for them in orbit.


However we faced several problems such as light, jet lag effects and sleep troubles.


For the well-being of the astronauts, the light exposure is a real struggle. As we can see from the already existing CBT program and the different types of lightning used in space stations, like GLAs or SSLAs, the solutions are not enough.

We propose to add to this already existing system an AI that would adjust astronauts' activities and will configure asynchronous teams.

Since it could detect the start of your "restoring" sleep, it could also adjust your schedule by taking into account the activities' priority and the well-being of your sleep cycle.


Now to the jet lag effect.

From a country to another, the body needs to adjust its biological clock to a stable rhythm, so we thought it would be the cause of the rupture of the astronauts' circadian rhythm.

However we could find on JAXA website (in the article "Tips for a Healthy Sleep Learned from Space Medicine") that the jet lag effect on space doesn't affect the astronauts sleep structure.

"Surprisingly, the brain waves of astronauts measured during their sleep in space show no major changes in sleep structure and biorhythm, while few reports have analyzed sleep in space in detail. Astronauts are given their own daily life schedule, which properly programs and manages their work and sleep schedules as well as daily physical exercise"

Therefore, the real two situations in which to readjust an astronaut sleep schedule would be the first travel of the crew member from their country to the launching station, and their sojourn in orbit.

Like we talked about when seeing the sleep cycle, we thought about the jet lag caused from their flight, and how we can use it for any other user. That's why we are proposing a jet-lag option that can reduce the jet lag by proposing sleep times correlated with your previous sleep habits to stabilize your sleep cycle.

Project Demo

video : https://www.youtube.com/watch?v=iMSDCm6p2WU&feature=youtu.be&fbclid=IwAR003gwv94Z_afux317yTXAXOdhg0p8Uusqb-2AZtidGxcYzLih_OwA_Xo4&ab_channel=Kamile


live project : https://leosheecool.github.io/website-hackathon-Space-App-Challenge2020/index.html

Data & Resources

https://pdfs.semanticscholar.org/4924/7ca3b385210d4ad75a106047fd280e01ca9e.pdf?_ga=2.151832889.1808507192.1593097374-1473586720.1593097374


https://www.researchgate.net/publication/229424960_Progress_in_our_understanding_of_the_architecture_of_paradoxical_sleep_since_William_Dement_and_Michel_Jouvet


On the importance of the uninterruption of REM sleep:

https://medweb4.unige.ch/labnic/papers/Oudiette-et-al_Sleep2012.pdf


On the relation between Mental fatigue and Cognitive functionning

https://www.longdom.org/open-access/evaluation-of-the-mental-fatigue-scale-and-its-relation-to-cognitive-and-emotional-functioning-after-traumatic-brain-injury-or-stroke-2329-9096.1000182.pdf

Tags
#sleep #optimization #schedule #smartobject
Judging
This project was submitted for consideration during the Space Apps Judging process.