A methodology for detecting and exploring non-convulsive seizures in patients with SAH

TL;DR

This paper introduces a new EEG-based methodology combining power spectrum estimation and empirical orthogonal functional analysis to detect and explore non-convulsive seizures in patients with subarachnoid hemorrhage, facilitating easier seizure identification.

Contribution

The novel approach simplifies seizure detection by transforming EEG signals into a multivariate time series analyzed through empirical orthogonal functions, improving detection accuracy.

Findings

Enables easier identification of non-convulsive seizures

Provides a systematic approach for seizure analysis

Enhances understanding of seizure patterns in SAH patients

Abstract

A methodology for understanding and de- tecting nonconvulsive seizures in individuals with sub- arachnoid hemorrhage is introduced. Specifically, begin- ning with an EEG signal, the power spectrum is esti- mated yielding a multivariate time series which is then ana- lyzed using empirical orthogonal functional analysis. This methodology allows for easy identification and observation of seizures that are otherwise only identifiable though ex- pert analysis of the raw EEG.