Federico, PaoloSpring, Aaron Michael2016-06-172016-06-1720162016http://hdl.handle.net/11023/3062High frequency oscillations (HFOs), characteristic oscillations observable using electroencephalography (EEG), are a promising and specific marker of the epileptogenic zone (EZ). However, there remain several obstacles to the implementation of HFOs as a prospective tool in the treatment of epilepsy. The identification of HFOs lacks universal standards and demands large time commitments from epileptologists. Therefore, this work involved the implementation and validation of novel frameworks for identifying HFOs, and for evaluating these HFOs as markers of the EZ. An epoched framework was implemented to facilitate the visual identification of HFOs, through which poor reliability was observed between reviewers. Furthermore, it was found that the temporal efficiency of visually evaluating HFOs within the epoched framework marked a substantial improvement over previously reported evaluation times. Using generalizability theory, it was then extended to determine effective methods of achieving highly reliable visual HFO evaluations, which included averaging ratings from at least 12 visual reviewers, or employing a training paradigm to increase the correlation of the ratings across reviewers. A novel surrogate marker (uGIC) of HFO activity was implemented and found to be correlated with HFOs detected algorithmically at low or high thresholds, all of which were found to be markers of the seizure onset zone overall. A retrospective framework was implemented to improve the accuracy of delineating the margins of resection with respect to EEG electrodes. A deformable method of image co-registration and a hybrid method of estimating electrode shift were shown to more effectively compensate for post-surgical shifts in brain anatomy. Resecting the uGIC or detected HFOs were found to have a positive effect on seizure freedom. Notably, the inclusion of lower-threshold oscillations was beneficial in facilitating the visual evaluation of HFOs, and in the identification of the EZ. Together, the findings of the studies undertaken herein provide a comprehensive framework to serve as the basis for transitioning high frequency oscillations to a feasible and meaningful part of the routine pre-surgical work-up in patients with focal epilepsy.engUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.NeuroscienceHealth SciencesMedicine and SurgeryEpilesyGeneralizability TheoryHigh Frequency Oscillations (HFOs)Electroencephalography (EEG)Intracranial EEGSurgeryInterrater ReliabilitylocalizationBiomarkersSurrogate MarkersVisual ReviewAutomated Detection Algorithmdoctoral thesis10.11575/PRISM/25954