Image still of Daniel Payne

[Video] How do weather, sleep, and time of day affect seizure risk forecasting?

Seer Medical is a research-driven medical technology company with a mission to empower people to take better control of their health.

Research Review is a series where we pull apart peer-reviewed journal articles published by Seer Medical’s research team. Building on our expertise in epilepsy, we are creating end-to-end systems that give people control over their health and put them on a path to better medical outcomes.

In December 2020, Epilepsia published Identifying seizure risk factors: A comparison of sleep, weather, and temporal features using a Bayesian forecast. This study expanded on previous works dedicated to seizure forecasting. It found that by using additional features, such as sleep, weather, and time of day, the accuracy of gauging seizure risk could be greatly improved.

Dr Daniel Payne is a data scientist on Seer Medical’s research team. His PhD explored how to design a seizure forecaster. Daniel joined the Seer Medical family with a focus on refining the seizure Risk feature on the Seer app. The seizure Risk feature will be available to users of the Seer app, providing them with the ability to gain greater insight into when their seizures are more or less likely to occur.

We caught up with Daniel recently to discuss:

  • What “Identifying seizure risk factors: A comparison of sleep, weather, and temporal features using a Bayesian forecast” is about.
  • The uncertainties people living with epilepsy face and how we can alleviate them.
  • Why Daniel was attracted to epilepsy research and what he is working on currently.
  • The difference between seizure forecasting and seizure prediction.
  • How to get the most accurate seizure risk gauge.
  • Features he believes could further improve seizure risk forecasting.


“Hi, my name is Daniel Payne, and I’m a data scientist at Seer Medical.”

What article will you be reviewing?

“The paper I’m reviewing today is “Identifying Seizure Risk Factors: A Comparison of Sleep, Weather, and Temporal Features Using a Bayesian Forecast.” It was published in 2020 in the journal Epilepsia.”

What is the article about?

“So the article is about using weather and sleep and time of day to forecast seizure risk.”

How does the article help people living with epilepsy?

“So people who have epilepsy live with uncertainty because they don’t know when their next seizure is going to occur. And we want to try to alleviate the uncertainty by providing risk forecasts for their seizures.”

Which feature had the biggest impact on improving seizure risk forecasting?

“The most impactful feature is time of day, and I think that’s because we’ve shown that there’s strong cycles throughout the day where you have low or high risk of a seizure. For example, some people don’t have seizures during sleep.”

What attracted you to epilepsy research?

“So, I actually had a seizure myself when I was young, and that sparked interest in neurology generally, and then, the opportunity came along when I started my PhD to work in epilepsy and I grabbed it.”

What are you currently working on at Seer Medical?

“So at the moment I’m working on the seizure risk gauge, which is about giving a forecast for your seizure risk, and that will work through the Seer app.”

Seizure forecasting vs. seizure prediction. What’s the difference?

“In seizure prediction, you’re either told you will have a seizure, or you won’t have a seizure, whereas forecasting gives you a risk, sort of like the weather forecast, where you’re told there’s an 80% chance of rain today instead of being told it will rain today.”

How can a person improve the accuracy of their seizure risk gauge?

“When we’ve put the seizure risk gauge into the app and have it working, the best way you can use it to get the best forecast is to record everything. So, record every seizure that you can, record everything about the seizure as well, and that will give us the best information to forecast your seizures.”

What other features could improve the accuracy of a seizure risk gauge?

“So heart rate is a factor that we want to look at, and it has shown a lot of promise. I am also personally quite interested in looking at cortisol, which is the hormone that measures stress because I think stress has a big impact on seizures as well.”