Is segmenting ECG features a viable tool for seizure detection?

Authors: Ewan S. Nurse and Mark J. Cook
Neurophysiologie Clinique Standard

Abstract

This article summarises new research on using ECG signals for seizure detection. While traditional methods rely on heart rate and variability, recent work shows that segmenting the full ECG waveform can capture richer cardiac changes linked to seizures. In a study of 31 patients, this approach achieved high accuracy (AUROC 0.91, 88% sensitivity, 83% specificity), suggesting ECG could serve as a practical alternative or complement to EEG-based detection. Though false alarm rates remain a challenge, advances in ECG analysis and wearable technologies highlight strong potential for real-world seizure monitoring.