Epileptic seizures don’t always last very long. And it is not necessarily the seizure itself that does the harm – it can be the uncertainty. If you don’t know when one of them will strike, it may prevent you from going for a swim or driving your car. It can affect your job – it can impact your whole life.
Seizures can, at first glance, seem quite random. But one of our big research findings has been to detect long-term risk patterns by looking at long periods of data. In my work, we go back to people’s entry logs for many years and then integrate the information we have gathered from their portable devices. We can then look at the combinations of factors that lead to periods of high risk.
We believe that for people with epilepsy, many biological rhythms can combine to create conditions conducive to seizures. It’s kind of like how several factors can combine to lead to a very high risk of bushfires in Australia. We know there are patterns in our climate – seasonal changes, changes over many years, sudden changes. All of these influences can combine to create conditions where extreme weather events are more likely.
It’s the same with seizures. We can see this combination of different factors that lead to periods of high risk for seizures. We can now give people an indication in advance of when they are entering a high-risk phase of their cycle, and, just as important, a low-risk phase – so they can plan for when they might be taking steps. vacation, for example.
I have a user who uses our app who has a cycle (approximately) of 21 days for their own risk of seizure. He wrote to me to say, âI was on vacation with the grandchildren, and it was so stressful to be with these grandchildren all the time. Fortunately, I was not in my high risk condition or I am sure I would have had a seizure.
It is very rewarding to hear a patient use this knowledge to make plans.
My job is to integrate data from multiple sources that record many types of body measurements and combine all of this information to help people understand their seizures.
The first research was done on a cohort of people who had a device implanted on the surface of the brain, just below the skull. It’s pretty invasive – it’s no longer the case. We are testing a new, less invasive device (Epiminder); it sits between bone and skin and records brain activity. It’s a fantastic development, but it’s still in the clinical trials stage.
What really excites us right now is this ability to infer people’s seizure risk rhythms simply by using a portable device – so it’s a completely non-invasive signal – combined with their own recording of their past seizures. . We just used a Fitbit, a consumer smartwatch. The signal we are most interested in is heart rate, but there is a whole range of relevant signals: oxygen saturation, electrodermal activity – this is the sweat sensor. Also skin temperature, quality of sleep, and exercise levels.
The risk factors can be very different for different people, so when you relate this data to the individual’s recorded history, it becomes suitable for them. The ânext big thingâ is really that patients just put their Fitbit on, look at the app on their phone, and get as much data and information as they got on any of the previous devices. You look at your risk factors in the morning the same way people look at the weather forecast for the week ahead.
Of course, there are limits. Capturing a person’s seizure history is much more accurate when we have the implant device, but it is something that is not yet available – and probably never will be for everyone, especially in developing countries. Epilepsy is a big deal, and with mobile and portable and remote cloud computing, we are actually able to deploy this technology in a scalable way that is very popular. It is power.
If you had asked me in high school what my dream job was, it was probably a science journalist writing for a magazine like Cosmos. I’ve always been interested in medicine, biology, physics and math, but I also loved English – writing was probably my favorite subject. But I was a shy kid and I think I realized how much interpersonal interactions there are in journalism. So I continued to follow the maths course in engineering, while keeping a very strong interest in medicine. Biomedical engineering was perfect.
In fact, I went through the University of Melbourne the very first year the degree was offered. âMedtechâ may have become a word, but there weren’t a lot of medical technology companies in Australia. The great name of the day was Cochlear. Beyond that, I didn’t know much about the industry. I had college mates who ended up going into accounting firms, and that didn’t interest me. So I did a doctorate just to stay in the field I love.
By the time I finished my PhD, there were so many more medical tech start-ups doing all kinds of wonderful things with mobile devices, apps, and wearable devices. I became part of the Seer team, which specializes in the management of epilepsy from the diagnostic stage to the lifelong management journey. They have developed what we call clinical grade portable devices, which means devices that record signals at the same level you would expect from a hospital grade device, but can be used at home. Today that includes consumer clothing like the Fitbit, which I use in my research and with the Seer app. We’re also working with companies like Epiminder, which make devices that can be implanted in the brain. That’s the whole scope.
I do less computing and data programming than I would like now. The further you advance in your research career, the greater the demands for funding, high-level ideas and strategy, so you pass the baton to doctoral students and other research assistants. And I have a lot of support from our industry partner, Seer – they have software developers and user experience researchers who are really helping to accelerate this clinical translation. But when I can, I like to enter the code.
The most exciting part of my job these days is finding new results from the data. We have found that there are some very interesting long term heart rhythms that seem to affect people’s likelihood of seizures. At the time these results were released, we were already piloting this signal into a forecast in the mobile app with the same participants. This kind of turnaround is incredible in terms of translating the academic research into technology that is currently available. It’s sort of unheard of.
As told to Graem Sims for Cosmos Weekly.
Do you like the weekly? You will love the COSMOS quarterly magazine.
Biggest news, in detail, quarterly. Buy a subscription today.