Date of Award
Doctor of Philosophy (PhD)
Epilepsy is a chronic neurological disorder characterized by seizures. Affecting over 50 million people worldwide, the quality of life of a patient with uncontrolled epilepsy is degraded by medical, social, cognitive, and psychological dysfunction. Fortunately, two-thirds of these patients can achieve adequate seizure control through medications. Unfortunately, one-third cannot.
Improving treatment for this patient population depends upon improving our understanding of the underlying epileptic network. Clinical therapies modulate this network to some degree of success, including surgery to remove the seizure onset zone or neuromodulation to alter the brain's dynamics. High resolution intracranial EEG (iEEG) is often employed to study the dynamics of cortical networks, from interictal patterns to more complex quantitative features. These interictal patterns include epileptiform biomarkers whose detection and mapping, along with seizures and neuroimaging, form the mainstay of data for clinical decision making around drug therapy, surgery, and devices. They are also increasingly important to assess the effects of epileptic physiology on brain functions like behavior and cognition, which are not well characterized.
In this work, we investigate the significance and trends of epileptiform biomarkers in animal and human models of epilepsy. We develop reliable methods to quantify interictal patterns, applying state of the art techniques from machine learning, signal processing, and EEG analysis. We then validate these tools in three major applications: 1. We study the effect of interictal spikes on human cognition, 2. We assess trends of interictal epileptiform bursts and their relationship to seizures in prolonged recordings from canines and rats, and 3. We assess the stability of long-term iEEG spanning several years. These findings have two main impacts: (1) they inform the interpretation of interictal iEEG patterns and elucidate the timescale of post-implantation changes. These findings have important implications for research and clinical care, particularly implantable devices and evaluating patients for epilepsy surgery. (2) They provide an analytical framework to enable others to mine large-scale iEEG datasets. In this way we hope to make a lasting contribution to accelerate collaborative research not only in epilepsy, but also in the study of animal and human electrophysiology in acute and chronic conditions.
Ung, Hoameng, "Mining Biomarkers Of Epilepsy From Large-Scale Intracranial Electroencephalography" (2016). Publicly Accessible Penn Dissertations. 2618.