Seizure forecasting

Seizure forecasting

Seizure risk forecasting can help the 30% of people living with drug-resistant epilepsy.

Why forecast seizures?

Not knowing when a seizure will start or happen is a major challenge for those living with epilepsy

The uncertainty of seizures can put many people living with epilepsy in danger. Even ‘small’ seizures can result in serious injuries or accidents if they happen whilst doing everyday activities like driving or cooking. In cases where people with epilepsy can’t control seizures with medications, many have no other choice but to design their lives around this uncertainty. 

If there was a way for people to know when they are most at risk for a seizure to happen, they could have a new means to manage the condition and plan their days with confidence. 

Bringing back control

Around two-thirds of people with epilepsy can eliminate their seizures by taking medications, however, this still leaves approximately 30% of people living with unpredictable seizures

Even so, lifelong medication isn’t an ideal solution for everyone, as anti-epileptic medication side effects can include dizziness, confusion, nausea, and memory problems. Some people will even tolerate infrequent seizures to avoid these side effects. 

Much like we rely on weather forecasts to make decisions about our days and even plans far into the future, seizure risk forecasting can help people with epilepsy manage their activities, optimise their medication treatments, and reduce anxiety.

Building on seizure cycles

Humans are governed by cycles and seizures are no exception

Since the earliest documented cases of epilepsy, people have reported cyclic rhythms of their seizures. In ancient times, it was thought that the phases of the moon must influence seizure risk. Today, many people continue to report cyclic effects on their seizures.

Our earlier research confirmed that seizures occur in cycles and that these cycles are common for all people with all types of epilepsy.

The cause of cycles remains a mystery and there are many possibilities, including internal and metabolic factors, changes in behaviour (such as sleep quality, stress levels, or exercise), and even environmental causes like seasonal variations. It’s likely that there are different causes for different people’s cycles.

A seizure cycle can exist without a person noticing. A major reason is that seizures can be infrequent. Another reason is that lots of people have multiple, overlapping cycles. A combination of cycles can appear irregular, and we use specialised analytic techniques to extract the cyclic information.

Proportion of people with epilepsy who have seizure cycles. Karoly et al (2020)

Seizure cycles are common for people with all types of epilepsy

Measuring cycles of risk with the Seer app

Our ongoing research has established that apps can play an important role in forecasting seizures

The Seer app is an epilepsy management app where users can record seizures, medication and other information, like mood and sleep quality. In our previous studies, we measured the cycles of seizure risk and determined how reliable they are to monitor someone’s chance of having a seizure. In these studies, we used the seizure times recorded in the Seer app to detect two possible cycles: a ‘fast cycle’ and a ‘slow cycle’.

The fast cycle represents a short cycle of one or two days, likely driven by circadian rhythms of the body (i.e. sleep). The slow cycle could be one week or longer. We combined these two cycles to develop a score of seizure likelihood — low, medium or high — that represented the risk of having a seizure in the next hour.

It’s important to note that our research confirms that seizure cycles based on self-reported events (where we rely on a person to recognise they’ve had a seizure and record them in the app) are just as accurate as seizure cycles based on events recorded during long-term EEG measurements.

We know that seizures can be triggered by many risk factors – a bad night’s sleep, missed medication, alcohol, or other changes in routine. Even when someone is at the low risk time of their cycle these triggers might still push their brain over the edge.

Next steps

Risk forecasts are likely to become much more powerful with other sources of information from wearable devices and implanted devices

Measuring factors that affect seizure cycles are key to further improving risk forecasts. Changes in hormones, metabolic factors, level of stress, amount of exercise, and sleep quantity and quality all play a role in influencing seizure cycles. Our research has also established that there is a link between heart rate cycles and seizure cycles.

Many of these factors can be measured through wearable devices such as health trackers and smart watches. The data recorded by a wearable device can be used to improve our ability to measure people’s seizure cycles and their risk factors.

Seizure forecasting opens up the opportunity to coordinate with other therapies, such as deep brain stimulation, to the timing of when event risks are the highest. Guided by a seizure risk forecast, it could be possible to tailor the stimulation to when a seizure is most likely to occur. By doing so, the severity of the event may be decreased or, even better, the event prevented altogether. Having this ability opens the door to providing people with epilepsy a higher quality of life by providing more effective epilepsy management beyond relying on medication.

More research behind seizure cycles and forecasting