Skip to content

Sleep science takes the stage at Big Data in Precision Health

Speakers at Stanford's Big Data in Precision Health conference discuss how their work with big data impacts and informs sleep research.

Sleep is one of the big black boxes of biology. We all know sleep is important and that our bodies need it to function, yet scientists are still in the dark when it comes to understanding how it works. 

That is starting to change, a group of scientists explained at Stanford's Big Data in Precision Health conference on Wednesday afternoon during a panel discussion on data-based approaches to sleep science.

Big Data in Precision Health, now in its seventh year, brings together leaders from academia, industry, government and medicine. It kicked off Wednesday morning with a welcome from Dean Lloyd Minor, MD.

"This is a really exciting time in data science," said Minor. "We're truly covering the gamut of issues related to precision health and biomedicine. In the next few days, we'll have speakers who will discuss everything from the social determinants of health, to how we can use machine learning and other analytic techniques to improve the drug discovery process."

During the sleep session, Jonathan Berent, a director at X, Alphabet's Moonshot Factory, described his focus on a particularly hot sector of sleep research: wearable technologies. Think watches, head bands, rings, even sensors embedded in a mattress to monitor sleep.

Although there's been quite a bit of evolution in this realm, sleep-monitoring technologies could benefit from some further finessing, Berent said, and in doing so they should prioritize something called a "daytime score." The benefit of sleeping is to rejuvenate yourself for the next day, but current technologies mostly focus on yielding data from the previous night's sleep. What could be more useful, said Berent, is a technology that harnesses data to tell users how prepared he or she is for the next day. And Berent said he thinks it will take more than just the monitoring devices themselves to achieve this.

"I think we can get an accurate 'daytime score' based on smartphone usage," he said. "We use our smartphone 50 or 60 times a day, sometimes a lot more than that, and that usage gives us a lot of data to draw statistics from." For example, there could be correlations between type or text response time and sleepiness. Linking smartphone usage data with sleep, Berent said, will be key for the next generation of sleep wearables.

In a slightly different vein, Jennifer Kanady, PhD, clinical innovation lead for sleep at Big Health, a digital mental health company, focuses on helping those who struggle to catch enough shut-eye. While many turn to pharmaceutical aids, Kanady has worked with her team to develop a different approach. The group has created an app rooted in cognitive behavioral therapy, which focuses on changes to a user's actions, thoughts and behavior to solve a problem.

The app employs tools such as sleep diaries and provides platforms for users to report their own sleep measures. There's even a sleep coach dubbed "the prof," a scholarly-looking cartoon with a delightful Scottish accent. The prof acts as a sort of in-app sleep guru, imbued with evidence-based knowledge and tips that help users get better sleep. 

"The field of sleep medicine is very large," said Emmanuel Mignot, MD, PhD, professor of psychiatry and behavioral sciences. "And the talks we've heard exemplify that: We can approach sleep medicine at home with an app, or we can create more devices to monitor sleep, but I think the first question we should ask is, 'Why should we study sleep?'"

Sleep problems are fairly common and vary widely, from sleep apnea and insomnia to restless leg syndrome, narcolepsy, and much more. "But what's really exciting to me is that sleep is at the center of a revolution of sorts," he said.

Alongside the development of new hardware and apps that keep tabs on sleep are new machine learning models. Researchers are using these models to analyze sleep quality objectively and to better understand sleep patterns. Mignot said he is particularly excited about the use of genomic and molecular data to better pin down the roots of certain sleep disorders. 

For example, Mignot said, analyses showed that many of the genes associated with narcolepsy (which causes extreme sleepiness and bouts of unexpected sleep) are actually involved in the immune system, which led to the knowledge that narcolepsy is an autoimmune disease. Outside of genomics, Mignot is hopeful that proteomics, or the study of proteins, will also illuminate new findings in sleep science.

"It's really incredible because, we can, for example, find proteins that are known to peak in the blood at very specific circadian times," said Mignot. By measuring hundreds of proteins that are associated with sleep and circadian rhythm, scientists can start to stitch together a clearer understanding of the molecular biology behind sleep cycles.

Now Mignot has teamed up with collaborating sleep labs to conduct a study in which they record sleep data from 30,000 participants.

At the end of the talk, a scientist in the audience stood to ask a question: What does sleep do? What does it accomplish?

"That's the million dollar question," said Mignot. "And that's what we hope to figure out with this study."

The conference continues today. Everyone is welcome to take part via webcast, by joining the conversation online using #BigDataMed, and by following our @StanfordMed feed for live tweets of keynotes and other proceedings from the conference. Scope will also feature additional stories about the event in the coming days.

Photos by Rod Searcey, Jonathan Berent shown at top

Popular posts

Animal Research
Could the avian flu be our next pandemic threat?

What does it mean that H5N1 bird flu, also known as highly pathogenic avian influenza A, is spreading among dairy cows? And how should U.S. health systems — and consumers of milk products — be responding?