Can ECG monitoring identify seizures?

J Electrocardiol. 2015 Nov-Dec;48(6):1069-74. doi: 10.1016/j.jelectrocard.2015.08.020. Epub 2015 Aug 6.

Abstract

Background: Seizures affect the autonomic control of the heart rate and respiration, and changes in these two variables are known to occur during, and even before the EEG onset of the seizure.

Goal: This work aims to quantify these changes and use them to identify the ECG onset.

Methods: Single-lead ECG signals were recorded from patients suffering from focal and generalized seizures. Two algorithms are proposed: one quantifies changes in the QRS morphology using principal component analysis, and one assesses cardiorespiratory interactions using phase rectified signal averaging.

Results: Positive predictive values of 86.6% and 77.5% and sensitivities of 100% and 90% were achieved for focal and generalized seizures respectively.

Conclusion: Results for focal seizures are in accordance with the literature, and detection of generalized seizures is improved after including respiratory information.

Significance: These findings could improve monitoring systems in epilepsy, and closed-loop techniques that aim to stop seizures.

Keywords: Autonomic; Childhood; Epilepsy; Respiration; Seizures.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Algorithms
  • Child
  • Child, Preschool
  • Computer Simulation
  • Diagnosis, Computer-Assisted / methods*
  • Electrocardiography / methods*
  • Female
  • Heart Rate*
  • Humans
  • Male
  • Models, Biological*
  • Reproducibility of Results
  • Respiratory Mechanics
  • Seizures / diagnosis*
  • Seizures / physiopathology*
  • Sensitivity and Specificity