Imaging the Epileptic Brain via High-frequency Oscillations from Multiscale Electrophysiological Analysis

About 23 million patients with epilepsy are refractory to medications and may require surgical resection of the epileptogenic zone (EZ) to control seizures. Currently, there is no direct measure available to locate the EZ. High-frequency oscillations (HFOs) has shown promise as a biomarker for localizing the epileptogenic brain. However, the utility and translation of HFOs are hindered by challenges in identifying and distinguishing pathological HFOs from non-epileptiform high-frequency activities, as well as localizing the epileptic tissue using HFOs. In this dissertation, the primary goal of this dissertation is to develop a feasible method for identifying and imaging HFOs that can accurately locate focal epileptogenic tissues and map the underlying epileptic network. The long-term objective is to gain a deeper understanding of the role and significance of HFOs in the structure, formation, and dynamics of the epileptic network, and ultimately to facilitate successful neural interventions for epilepsy practices. Towards this goal, this study has (i) developed a data-driven method to automatically identify pathological HFOs from nonepileptiform activities; (ii) investigated the characteristics of HFOs and epileptic spikes in the presence or absence of the cooccurrence status; (iii) demonstrated the utility of the identified HFOs in delineating the cortical regions of epileptic abnormality validated by clinical evidence of surgical resection and SOZ, (iv) investigated the cortical HFOs in association with the pathological evidence and establishing the reliability of HFOs in the underlying epileptic network. Together, the proposed approach has been systematically evaluated with simulations and clinical studies. Overall, the results show morphological and source imaging evidence that pathological HFOs can be identified and utilized to delineate the underlying epileptogenicity, especially in patients with multiple spike variations. Further analysis also suggests that the underlying generation of HFOs involves a stable and repeatable process with individual variations. This work demonstrates the potential of interictal HFOs in localizing the epileptic brain, providing a promising translational tool for the management of epilepsy in vulnerable patients.