WEDNESDAY, DECEMBER 10, 2025 (HealthDay News) — Artificial intelligence (AI) can help improve epilepsy treatment by connecting the dots in complex cases, according to a new study.
AI helped identify patients with drug-resistant epilepsy who could benefit from surgery and exposed missing tests and assessments that could better guide patient care, researchers in Atlanta reported Friday during a meeting of the American Epilepsy Society.
“Patients see multiple specialists and undergo numerous tests, and their treatment plans evolve over the years, so it is easy to miss important details or opportunities for surgical evaluation,” said lead researcher Dr. P. David Adelson in a press release. He is vice president of the Rockefeller Neuroscience Institute at West Virginia University in Morgantown.
“Large-language AI models continuously read, organize and interpret medical records to help healthcare teams stay one step ahead,” said Adelson. “They have the potential to transform not just epilepsy care, but any long-term condition that requires cross-disciplinary coordination.”
For the study, researchers used an AI to evaluate the records of 820 epilepsy patients who made appointments with neurologists at WVU Medicine.
The AI identified 88 patients (11%) who met criteria for drug-resistant epilepsy, according to the researchers. Guidelines recommend that these people be evaluated for surgery, although not everyone is ultimately a candidate for it.
The AI also found that 54% of patients had outdated or missing MRIs; Neuropsychological examinations were missing in 91%; and 35% had missing electroencephalograms (EEGs).
Closing these gaps could lead to shorter delays in treatment and fewer missed opportunities for patients, Adelson said.
“It’s like having an intelligent research assistant that never gets tired. It ensures that the progress of care is coordinated and complete, making specialists more efficient by ensuring that no patient is missed due to information gaps or time constraints,” he added. “This allows specialists to focus on interpreting results and planning treatment, helping to extend expert-level care to more patients, even in hospitals with limited specialty coverage.”
Epilepsy care is complex and often fragmented and represents a major opportunity for AI, said Dr. Howard Goodkin, former president of the American Epilepsy Society, in a press release.
“Research like this demonstrates the growing potential of responsible artificial intelligence to improve epilepsy care,” said Goodkin, who was not involved in the study. “Rather than replacing the epilepsy specialist, AI acts as a partner, improving the human experience in medicine by tracking complex medical information, identifying gaps and stimulating action to ensure care stays on track over the long term.”
Results presented at medical meetings should be considered preliminary until published in a peer-reviewed journal.
More information
The Epilepsy Foundation has more information on drug-resistant epilepsy.
SOURCE: American Epilepsy Society, press release, December 5, 2025
About The Author
