PITTSBURGH – Clinician-scientists at the University of Pittsburgh School of Medicine have developed an artificial intelligence (AI) tool that significantly improves the diagnostic accuracy of functional seizures, which are often misdiagnosed as epilepsy. The findings, published today in the journal Epilepsia, demonstrate how AI can support non-neurology clinicians by flagging complex cases for further consideration, with the potential to reduce unnecessary drug use and improve patient safety and access to expert diagnostic care.
“Even for highly trained clinicians, it is difficult to distinguish between functional and epileptic seizures. Most seizures are epileptic, so clinicians start there, but the remaining cases represent a significant gap in treatment,” said lead author Wesley Kerr, MD, assistant professor of neurology and bioinformatics at Pitt and chief epileptologist in the UPMC Functional Neurological Disorders Program. “AI is not about replacing clinicians; it is about extending our impact by identifying complex cases faster and making care more compassionate and personal.”
Functional seizures affect an estimated 10% of people with seizures. Unlike epileptic seizures, which are caused by electrical activity in the brain and can be managed with appropriately selected drugs or surgery, functional seizures are caused by a combination of biological, psychological, and social factors. Because seizures are not caused by abnormal electrical activity in the brain, antiseizure drugs are ineffective. Instead, treatment focuses on neurobehavioral therapy to help patients and clinicians understand the underlying stressors that push the body into “panic mode.”
“Functional seizures are ‘panic without panic,’ which is the body’s response to feeling overwhelmed,” Dr. Kerr says. “Your body is having a panic attack, but you don’t feel it because your brain protects you from the sensation. Hidden stress can explode as an attack.”
To understand whether AI tools can effectively support clinician decision-making, the researchers presented 117 anonymized patient cases to 163 reviewers with varying backgrounds and experience levels, from non-clinicians, medical students, non-neurologists to epileptologists. Each reviewer evaluated a randomized subset of cases, and their accuracy was measured before and after AI assistance.
The analysis found that AI tools improved diagnostic accuracy for 66% of reviewers. Among average AI-savvy users, the accuracy of diagnosing functional seizures improved by nearly 20%. Notably, expert clinicians, including most epileptologists, did not show significant improvement as the AI largely replicated existing knowledge.
“AI won’t make us better professionals, but it will certainly give clinicians who have never heard of functional seizures the tools to say, ‘Think about this,’ so we can improve care for everyone, no matter where they live,” Kerr said.
As an immediate next step for this research, Professor Kerr is exploring ways to incorporate AI-assisted analysis of video-recorded seizures, which could reduce the need for inpatient monitoring, the current gold standard for confirming the cause of seizures.
Kerr and his team are now moving toward the goal of being able to automatically monitor and flag potential cases of functional seizures within a patient’s health chart. Scientists hope that additional AI-assisted checkpoints, combined with clinical oversight and ethical considerations, will help identify people who could benefit from a deeper understanding of seizures and help clinicians more quickly and accurately identify appropriate drug and non-drug treatments. They point out that reducing medications that do not treat the cause of seizures may be especially important for pregnant, elderly, and medically complex patients.
For more information about this research and seizure treatment at UPMC, please visit:
Patients who want to know whether their seizures may be functional should contact UPMC Epilepsy for an initial consultation.
Other authors of the study include Katharine McFarlane, MD, James Castellano, MD, Anto Bagich, MD, Rachna Reddy, Sara Yagoobi, MD, Laura Kirkpatrick, MD, Daniel Kearns, MD, Zongchi Xia, MD, and Paige Pennell, MD. All reside in Pitt.
This study was supported by the National Institute of Neurological Disorders and Stroke (K23NS135134, R25NS065723, R25NS089450, U24NS107158, R01NS033310, P20NS080131, T32GM08042, T90DA022768, R90DA022768, R90DA023422, K12NS098482), American Academy of Neurology, American Brain Foundation, Epilepsy Foundation, American Epilepsy Society, Epilepsy Research Consortium, William M. Keck Foundation, and Christina Louise George Trust.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or position, and all views, positions, and conclusions expressed herein are solely those of the authors. Read the full text here.
