Electrocorticography and Stereo EEG provide distinct measures of brain connectivity: implications for network models

Brain network models derived from graph theory have powerful potential to guide functional neurosurgery, and to improve rates of post-operative seizure freedom for patients with epilepsy. A barrier to applying these models clinically is that intracranial EEG electrode implantation strategies vary by center, region and country, from cortical grid & strip electrodes, to purely stereotactic depth electrodes, to a mixture of both. To determine whether models derived from one type of study are broadly applicable to others, we investigate the differences in brain networks mapped by electrocortiography (ECoG) and stereoelectroencephalography (SEEG) in a matched cohort of patients who underwent epilepsy surgery. We show that ECoG and SEEG map broad network structure differently, and demonstrate substantial disparity in the ability of node strength to localize the epileptogenic zone in SEEG compared to ECoG. We demonstrate that eliminating white matter contacts and reducing network nodes to anatomical regions of interest rather than individual contacts improves the ability of these models to distinguish between epileptogenic and non-epileptogenic regions in SEEG, but this distinction is hampered when implants do not sample a high portion of local, intralobar connections. Our findings suggest that effectively applying computational models to localize epileptic networks requires accounting for the effects of spatial sampling, particularly when analyzing both ECoG and SEEG recordings in the same cohort. Finally, we share all code and data in this study, aiming for our findings to accelerate research in brain network connectivity in epilepsy and beyond.