The Signal Energy of Seizures
Abstract number :
1.128
Submission category :
Year :
2001
Submission ID :
911
Source :
www.aesnet.org
Presentation date :
12/1/2001 12:00:00 AM
Published date :
Dec 1, 2001, 06:00 AM
Authors :
H.P. Zaveri, PhD, Department of Neurology, Yale University, New Haven, CT; R.B. Duckrow, MD, Department of Neurology, University of Connecticut Health Center, Farmington, CT; S.S. Spencer, MD, Department of Neurology, Yale University, New Haven, CT
RATIONALE: The study of a phenomenon by system theory is grounded in a characterization of the signal power of the observed time-series. This study was performed to characterize the signal energy, a measure related to signal power, of intracranially recorded seizures in patients with intractable epilepsy.
METHODS: Signal energy was measured by a model (simple harmonic motion) based algorithm that nonuniformly weighs the signal spectrum so that high-frequency signals contribute greater energy than low-frequency signals. A total of 38 seizures from 8 patients were studied. Six of these patients had medial temporal onset of seizures, three with mesial temporal sclerosis and three with paradoxical temporal lobe epilepsy. The other two patients had neocortical, temporal and frontal lobe, onset of seizures. Approximately 500 seizure profiles recorded from onset and spread locations were studied.
RESULTS: The energy profiles of most seizures revealed a single activation and only a few seizures (8%) had two or more distinct activations during the course of the seizure. Seizure energy profiles demonstrated either a rapid rise (33%) or a gradual rise (42%) in signal energy at seizure onset. In most seizures (75%) this was followed by either a steady decline to seizure end (53%) or an abrupt transition at the end of the seizure (22%). These energy profiles were observed at both onset and spread locations and at times occured with a high degree of correlation.
CONCLUSIONS: Focal seizures have a relatively simple signal energy profile with one or more activations followed by a decline from maximum. The energy profile of these seizures suggests that the main segment of a focal seizure may be viewed as a single event, one that manifests itself with distinct onset, sustenance and termination characteristics. The gradual decrease in signal energy that was frequently observed is suggestive of a dissipative (inhibited) system. The observed characterisics of seizure signal energy profiles and the high degree of correlation of these profiles also suggests that during seizures a distributed network is activated wherein one or more elements of the network paces the areas involved in the seizure.