Written by Julie Penfold, RARE Revolution insider

Estimated reading time: 6 minutes

EpiScalp is a new tool developed by researchers at Johns Hopkins University. By harnessing the power of AI they aim to aid more accurate and timely diagnosis of epilepsies which affect over 70 million people globally

Epilepsies are common, yet serious neurological conditions estimated to affect more than 70 million people globally¹. In the UK, there are around 630,000 people living with a form of epilepsy², although the exact number is currently unknown.

That’s because misdiagnosis of epilepsy is common. NHS RightCare estimates the UK rate of misdiagnosis is around 20-30%. When patients experience delays in getting a correct diagnosis it can have many impacts, including medication side effects, unnecessary driving restrictions and other quality of life challenges.

But doctors could soon reduce epilepsy misdiagnosis rates by up to 70%. Researchers at Johns Hopkins University have developed a tool that can turn routine electroencephalogram (EEG) tests that appear normal into highly accurate epilepsy prediction tests.

The tool works by uncovering hidden epilepsy signatures in seemingly normal EEGs. The research team’s initial study found the tool could significantly reduce false positives, seen in around 30% of cases globally.

“Even when EEGs appear completely normal, our tool provides insights that make them actionable,” explains Sridevi V Sarma, a John Hopkins biomedical engineering professor who led the work. “We can get to the right diagnosis three times faster because patients often need multiple EEGs before abnormalities are detected, even if they have epilepsy. Accurate early diagnosis means a quicker path to effective treatment.”

How EpiScalp works

Epilepsy causes recurrent and unprovoked seizures triggered by bursts of abnormal electrical activity in the brain. Standard care involves analysing scalp EEG recordings during initial evaluations. Clinicians partly rely on EEGs to diagnose epilepsy and decide whether patients need anti-seizure medications following this testing.

However, EEGs can be challenging to interpret as they also capture noisy signals and seizures rarely occur during the typical 20-40 minutes of an EEG recording. “These characteristics make diagnosing epilepsy subjective and prone to error, even for specialists,” Sarma explains.

Sarma’s team studied what happens in the brains of patients when they are not experiencing seizures. The tool they have developed, EpiScalp, uses algorithms trained on dynamic network models to map brainwave patterns and identify hidden signs of epilepsy from a single, routine EEG.

One of the questions the researchers tried to get an answer for was—if a patient has epilepsy, why don’t they have seizures all the time? “We hypothesised that some brain regions act as natural inhibitors to suppress seizures. It’s almost like the brain’s immune response to the disease,” she suggests.

The team’s initial study analysed 198 epilepsy patients from five major medical centres: Johns Hopkins Hospital, Johns Hopkins Bayview Medical Center, University of Pittsburgh Medical Center, University of Maryland Medical Center, and Thomas Jefferson University Hospital.

From the 198 patients they analysed, 91 patients had epilepsy while the rest had non-epileptic conditions mimicking epilepsy. Following this, when Sarma’s team reanalysed the initial EEGs using the EpiScalp tool, it ruled out 96% of false positives. This would cut potential misdiagnoses among these cases from 54% to 17%, they say.

“This is where our tool makes a difference because it can help us uncover markers of epilepsy in EEGs that appear uninformative, reducing the risk of patients being misdiagnosed and treated for a condition they don’t have,” says Khalil Husari, co-senior author of the study and assistant professor of neurology at Johns Hopkins University.

“These patients experienced side effects of the anti-seizure medication without any benefit because they didn’t have epilepsy,” he adds. “Without the correct diagnosis, we cannot find out what is actually causing their symptoms.”

Understanding misdiagnosis causes

In certain cases, researchers found misdiagnosis happens due to misinterpretation of EEGs. This can result in doctors potentially over-diagnosing epilepsy to prevent the dangers of a second seizure, the study suggests.

Additionally, the study found in some cases, patients experienced nonepileptic seizures, which mimic epilepsy. These conditions can often be treated with therapies that do not involve epilepsy medication.

In earlier work, the team studied epileptic brain networks using intracranial EEGs to demonstrate that the seizure onset zone is being inhibited by neighbouring regions of the brain when patients are not seizing. EpiScalp builds on this research, they say, by identifying these patterns from routine scalp EEGs.

The study, published in the journal Annals of Neurology, found traditional approaches to improve EEG interpretation often focus on individual signals or electrodes. The EpiScalp tool instead analyses how different regions of the brain interact and influence one another through a complex network of neural pathways.

“If you can look at how nodes are interacting with each other within the brain network, you can find this pattern of independent nodes trying to cause a lot of activity and the suppression from nodes in a second region, and they’re not interacting with the rest of the brain,” explains Patrick Myers, first study author and doctoral student in biomedical engineering at Johns Hopkins.

“We check whether we can see this pattern anywhere,” he adds. “Do we see a region in a patient’s EEG that has been decoupled from the rest of the brain’s network? A healthy person shouldn’t have that.”

The team is now conducting a larger prospective study to further validate its findings across three US epilepsy centres. It also filed a patent for the EpiScalp technology in 2023.

Promising research

Johns Hopkins University’s initial study is a cause for real optimism, says the UK’s leading charity for this condition, Epilepsy Action.

“It’s really promising to see new developments in technology that can potentially reduce chances of misdiagnosis,” says Tom Shilliton, Epilepsy Action’s health improvement and research manager.

“We’re very excited to see the outcomes of the larger US trial for this tool. This could help give solid evidence on the effectiveness of this new technology, and how it can potentially help people with epilepsy in the future.”

The charity says it hears of “many stories of misdiagnosis or late diagnoses” from their community. “Any new discoveries that can help improve this process are certainly encouraging,” Shilliton adds.

“We’re looking forward to hearing more about the trial outcomes and the practical applications of the EpiScalp tool.”

Connect with Sridevi

• LinkedIn

References
[1] https://pubmed.ncbi.nlm.nih.gov/30686584/
[2] https://www.epilepsy.org.uk/info/what-is-epilepsy

innovation spotlights the future of RARE through a technological lens. To access more innovation articles click below.