Video Technologies for Neonatal Seizures

Professor Nicolaos Karayiannis analyzes video recordings of newborn babies as he develops a new automated seizure detection system for neonates

Project Information

A quick and accurate diagnosis is paramount to detecting seizures in newborn babies because each seizure can damage the developing brain and the most effective anticonvulsive drug treatment depends on timely detection and assessment. Unfortunately, today clinicians must analyze visually lengthy data (records of the electrical activity of the brain called EEG) and watch hours of video recordings to identify seizures.

This project focuses on the development of automated video processing and analysis procedures aimed at the characterization and recognition of neonatal seizures. This is achieved by relying on recent developments in computer and video technology to extract, process and analyze novel quantitative information on the behavioral characteristics of neonatal seizures. The long-term goal is the development of a stand-alone automated system that could be used as a supplement in the neonatal intensive care unit to provide 24-hour-a-day, noninvasive monitoring of infants at risk for seizures.