Greater epileptiform activity burden is associated with worsened functional outcomes at discharge in acutely ill, hospitalized patients. Automated measures of continuous electroencephalogram (cEEG) data could provide useful prognostic insights in clinical research.
Why this matters
Epileptiform activity is defined as seizures and seizure-like periodic and rhythmic patterns of brain activity and is detected in up to 50% of critical care patients receiving cEEG monitoring.
The relevance of the burden of epileptiform activity to outcomes across a broad range of patients and conditions is unclear. An automated, machine learning approach to annotate and interpret cEEG data from hospitalized patients could be a useful tool for providing rapid prognostic advice.