Electronic Theses and Dissertations

Identifier

6330

Date

2018

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Engineering

Department

Electrical & Computer Engineering

Committee Chair

Mohammed Yeasin

Abstract

Epilepsy—a disorder that is far more common than is widely realized—results in high morbidity and even mortality. It is defined semiologically in part, but it is a disorder caused by the disturbed synchronization of natural brain oscillations. The current standard treatment is implanting intracranial electrodes that are continuously connected to an acquisition system while the patient waits in an Epilepsy Monitoring Unit (EMU) until the patient has a seizure. Given enough seizures, this information can be taken to the operating room. Then, the electrodes, which had shown pathologic activity, are marked and surgical resection of the determined pathologic areas follows. This entire process can take up to a month in any given patient and results in considerable patient and system costs. It is known that there are electrophysiologic markers that happen between seizures or interictally. However, the question of whether those markers can define the seizure onset zone (SOZ) adequately enough to perform resection has not been resolved completely yet. The purpose of this work is to explore those electrophysiologic biomarkers and define the methods to both detect them reliably and compare them to previously determined SOZ. First, high frequency oscillations (HFO)—a now heavily explored interictal electrophysiologic biomarker—are investigated via a pre-worked detector; its role in SOZ determination is considered in the context of both old (interictal epileptiform discharges) and new (phase-amplitude coupling) biomarkers. Further, work is explored for automating the localization process via a machine learning algorithm to automatically classify the SOZ and non-SOZ. We also compared the rate of HFO in/out of SOZ and the resection area in four different epochs: at night, awake time, preictal, and ictal. Seizures initiate when most or all neurons in epileptic regions start to fire synchronously. Evidence obtained from the entorhinal cortex (EC) in animal models of epileptiform synchronization show that low-voltage fast (LVF) onset seizures are initiated by synchronous inhibitory events. We sought to establish whether the increased firing of inhibitory interneurons occurs at the onset of spontaneous LVF seizures in patients with mesial-temporal lobe epilepsy, and whether the increased firing of excitatory neurons follows this.

Comments

Data is provided by the student

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses and dissertation (ETD) repository.

Download

Share

COinS