COLUMBUS, Ohio – In a new study, researchers at The Ohio State University used artificial intelligence (AI) tools to predict seizure outcomes in mouse models.
Seizures may cause temporary disruptions in motor control, which typically follow certain behavioral patterns that help identify seizure types, onset zones and potential outcomes.
Individuals who experience ongoing seizures have an increased risk of various complications, including physical injuries or sudden unexpected death in epilepsy (SUDEP).
Investigators with The Ohio State University Wexner Medical Center, College of Medicine and College of Veterinary Medicine explored the potential for AI-assisted tools to analyze complex behavioral patterns and distinguish seizure outcomes using data-driven techniques.
Study findings are published in the journal Annals of Neurology.
"Our AI-aided behavior analysis provides a powerful and objective approach to understanding seizure dynamics in a mouse epilepsy model. By analyzing movement patterns in a data-driven manner, we can accurately classify seizure types and predict outcomes," said study senior author Bin Gu, PhD, assistant professor with Ohio State's Department of Neuroscience. "This approach enhances accuracy and efficiency in epilepsy research and clinical practice, while opening new possibilities of personalized treatment strategies." Analyzing these behaviors in epilepsy is complex and presents ongoing challenges. Manual video inspection is subjective, time-consuming and can overlook key behavioral details and movement dynamics, said Gu, who also is a member of Ohio State's Chronic Brain Injury Program and Translational Data Analytics Institute.
Researchers used two AI-aided tools to decode underexplored behavior and action domains of induced seizures in 32 inbred mouse strains that mimic human genetic diversity, as well as in a mouse model of Angelman syndrome. They identified 63 interpretable behavior groups.
Analysis of these behavioral groups reveals distinct patterns that distinguish seizure states, track progression, and relate to mouse sex, strain, and mutations.
"Looking ahead, the ability to analyze videos captured on smartphones and home surveillance devices has exciting implications. This could transform how seizures are recognized outside of the clinic, giving families and caregivers a simple way to contribute valuable diagnostic information," said Gu. "This especially critical in the telemedicine settings, where access to neurologists and epileptologists is often limited, particularly for patients in underserved and rural communities."
Yuyan Shen, study first author and a graduate research assistant with Ohio State's Department of Neuroscience and a doctor of veterinary medicine candidate with Ohio State's College of Veterinary Medicine, noted that similar approaches could also benefit pets with epilepsy, where video analysis may help veterinarians and owners better diagnose and manage seizure activity in everyday environments.
/Public Release. This material from the originating organization/author(s) might be of the point-in-time nature, and edited for clarity, style and length. Mirage.News does not take institutional positions or sides, and all views, positions, and conclusions expressed herein are solely those of the author(s).View in full here.
