Authors :
Presenting Author: Elliot Nester, MS – Barrow Neurological Institute
Noah Hutson, PhD – Barrow Neurological Institute
Stephen Foldes, PhD – Barrow Neurological Institute
David Harris, MD – Barrow Neurological Institute
Leon Iasemidis, PhD – Barrow Neurological Institute
Andrew Yang, MD, MS – Barrow Neurological Institute
Rationale:
The clinical efficacy of deep brain stimulation (DBS) of the anterior nucleus of the thalamus (ANT) suggests a critical role of the thalamus in the propagation, termination, and perhaps even the generation of focal-onset seizures. Neuromodulation with DBS as well as responsive neurostimulation (RNS) are being increasingly utilized with thalamic nuclei other than ANT, such as pulvinar and centromedian (CM) nuclei. However, ictal electrophysiological dynamics throughout the thalamocortical network remain poorly characterized in the literature.Methods:
Stereo EEG (SEEG) data was obtained from 40 seizures (14 patients). The periodic and aperiodic components of the local field potential (LFP) were extracted using the ‘fitting oscillations and one over f’ (FOOOF) algorithm, which removes narrowband oscillatory signals from the LFP, isolating its aperiodic (1/f) component. The slope of this aperiodic signal (computed from 6 to 50 Hz) is a marker of the excitation-inhibition balance within the local neuronal circuit. The midpoint of the aperiodic component (between 1 and 150 Hz) corresponds to broadband power, which is a well-established correlate of local neuronal spiking.
Bipolar-referenced signals were obtained from the cortical seizure onset zone (SOZ), thalamus (12 ANT, 2 pulvinar), cortical regions immediately adjacent to the SOZ (near-SOZ), and cortical regions in anatomically homologous areas in the contralateral hemisphere (far-SOZ). The two LFP metrics were depicted across the peri-ictal window (-25% to +125% from onset), where time was normalized as percentage of seizure duration to control for variable durations across seizures/subjects. Seizure-level data were first z-score normalized with respect to corresponding baselines (< -25% from onset), and then averaged across seizures/subjects for group-level results.