Epilepsy is a condition that causes repeated, unexpected seizures due to unusual electrical activity in the brain. Doctors use scalp EEGs to diagnose it, which involve placing small electrodes on the scalp to monitor brainwave patterns.
Doctors use EEGs to help diagnose epilepsy and decide on treatment. However, EEGs can be tricky to interpret because they often capture a lot of extra noise, and seizures are rare during the short 20-40-minute recording. This makes diagnosing epilepsy challenging and sometimes inaccurate, even for experts.
This Johns Hopkins University study reveals that a new tool, called EpiScalp, could dramatically cut epilepsy misdiagnoses by up to 70%. The tool achieves this by identifying hidden epilepsy indicators that appear normal in routine EEGs.
This advancement could substantially reduce the global misdiagnosis rate, currently around 30%, and spare patients from the side effects of unnecessary medication, driving limitations, and other life disruptions due to incorrect diagnoses.
Sridevi V. Sarma, a Johns Hopkins biomedical engineering professor who led the work, said, “Even when EEGs appear completely normal, our tool provides insights that make them actionable. 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.”
Epilepsy is associated with brain volume and thickness differences
The team sought to improve the reliability of epilepsy diagnosis by examining brain activity during periods without seizures. Their tool, EpiScalp, uses advanced algorithms and dynamic network models to detect hidden signs of epilepsy in routine EEGs.
Sarma explained, “If you have epilepsy, why don’t you have seizures all the time? We hypothesized that some brain regions act as natural inhibitors, suppressing seizures. It’s like the brain’s immune response to the disease.”
The study involved 198 patients from five major medical centers: Johns Hopkins Hospital, Johns Hopkins Bayview Medical Center, University of Pittsburgh Medical Center, University of Maryland Medical Center, and Thomas Jefferson University Hospital. Among these patients, 91 had epilepsy, while the rest had conditions mimicking epilepsy.
After analyzing the initial EEGs, the team found that the tool ruled out 96% of those false positives, cutting potential misdiagnoses among these cases from 54% to 17%.
EpiScalp is instrumental in reducing epilepsy misdiagnoses by uncovering hidden markers in seemingly normal EEGs. This can prevent patients from being wrongly diagnosed and treated for epilepsy, sparing them from unnecessary medication side effects.
Incorrect diagnoses can delay finding the real cause of patients’ symptoms. EpiScalp improves diagnostic accuracy to help ensure patients receive the correct treatment for their condition.
In some cases, doctors may misdiagnose epilepsy due to EEG misinterpretation. To prevent the dangers of a second seizure, they might overdiagnose epilepsy. However, some patients experience nonepileptic seizures that mimic epilepsy and can be treated without epilepsy medication.
Previous research on epileptic brain networks using intracranial EEGs showed that neighboring brain regions inhibit the seizure onset zone when patients aren’t having seizures. EpiScalp builds on this by identifying these patterns in routine scalp EEGs.
Patrick Myers, the study’s first author and a doctoral student in biomedical engineering at Johns Hopkins, explained that, Unlike traditional methods focusing on individual signals or electrodes, EpiScalp examines how different brain regions interact through a complex network of neural pathways.
Patrick Myers, first author and doctoral student in biomedical engineering at Johns Hopkins, said, “If you just look at how nodes are interacting with each other within the brain network, you can find this pattern of independent nodes trying to cause much activity and the suppression from nodes in a second region, and they’re not interacting with the rest of the brain.”
“We check whether we can see this pattern anywhere. Do we see a region in your EEG that has been decoupled from the rest of the brain’s network? A healthy person shouldn’t have that.”
Journal Reference:
- Patrick Myers, Kristin M. Gunnarsdottir, Adam Li, Vlad Razskazovskiy, Jeff Craley, Alana Chandler, Dale Wyeth, Edmund Wyeth, Kareem A. Zaghloul, Sara K. Inati, Jennifer L. Hopp, Babitha Haridas, Jorge Gonzalez‐Martinez, Anto Bagíc, Joon‐yi Kang, Michael R. Sperling, Niravkumar Barot, Sridevi V. Sarma, Khalil S. Husari. Diagnosing Epilepsy with Normal Interictal EEG Using Dynamic Network Models. Annals of Neurology, 2025; DOI: 10.1002/ana.27168
Source: Tech Explorist
