Synchronization Analysis of Depth and Extracranial EEG Seizure Data in Subarachnoid Hemorrhage (P1.269)

OBJECTIVE:To identify dominant oscillators within and between depth and extracranially-recorded EEG data during non-convulsive seizures in SAH patients. BACKGROUND:Seizures are frequent after acute brain injury such as subarachnoid hemorrhage (SAH) and are associated with poor outcome, particularly if seizures do not generalize. Knowledge of the underlying oscillatory drivers that differentiate seizures that propagate from those that do not will be important to develop more targeted treatment strategies. Advanced quantitative analysis tools, particularly synchronization methods will be necessary to fully understand the nature of these recordings and in the future will aide patient physiology-driven choice and timing of interventions. DESIGN/METHODS:We included SAH patients that underwent invasive brain monitoring including a minidepth electrode taken from a database of SAH that has been compiled as a subset of the Subarachnoid Hemorrhage Outcomes Project. Matlab source code was developed to decompose each channel data using empirical mode decomposition and then calculate mean phase coherence and spectral estimation of each channel data. First, data was clipped and divided into groups of (a) surface-only seizures, (b) depth-only seizures, (c) depth and surface seizures and (d) interictal samples. These samples were then decomposed into oscillator sets using empirical mode decomposition. Frequency and phase measures were extracted using the Hilbert analytic method and mean phase coherence was calculated to obtain spectral and synchronization information. RESULTS:Underlying basis functions were visualized for all channels and synchronization levels plotted in three dimensional surface plots. Synchronization intensity and spatial organization of synchronization varies with different seizure types, particularly whether the seizure stays confined to depth electrodes or propagates to involve other brain regions and in the interictal phase. CONCLUSIONS:Information on synchronization may help delineate outcomes and targeted treatment protocols in these cases given that prognosis varies whether the seizure remains confined to depth electrodes or propagates. Study Supported by: National Library of Medicine - National Institutes of Health grant RO1 M06910 Disclosure: Dr. Fine has nothing to disclose. Dr. Albers has nothing to disclose. Dr. Schmidt has nothing to disclose. Dr. Pugin has nothing to disclose. Dr. Hripcsak has nothing to disclose. Dr. Claassen has nothing to disclose.