Advanced electrocorticography and its translation to epilepsy surgery

This thesis investigates how intraoperative electrocorticography (ioECoG) can be optimized to improve surgical outcomes in patients with drug-resistant focal epilepsy. Achieving seizure freedom depends on accurate delineation of the epileptogenic zone, which can be challenging. ioECoG tailoring can help delineate epileptic tissue in real-time during epilepsy surgery. Traditional tailoring relies on visual identification of interictal spikes and high-frequency oscillations (HFOs). However, interpretation remains subjective and spatial sampling is limited by conventional electrode grids.
Part I explores advanced signal analysis techniques to enhance objective identification of epileptic tissue. In patients with focal cortical dysplasia and favorable outcomes, combined spatial and temporal characteristics of spikes and HFOs – particularly ripples co-occurring with spikes – were strongly associated with resected tissue. Beyond event-based markers, we demonstrated that using spectral band powers subtle, visually imperceptible frequency patterns can distinguish epileptic from non-epileptic tissue. Furthermore, spectral entropy variability cannot only distinguish epileptic from non-epileptic tissue, but it can also predict post-surgical seizure outcome, which supports patient counseling.
Part II addresses clinical interpretation and innovation. We assessed optimal montages for spike and HFO visualization. We explored the spatial relationship between MRI lesions and electrophysiological abnormalities. Moreover, we implemented high-density ioECoG grids to improve spatial resolution. Together, these studies aim to support more precise, reproducible, and data-driven surgical decision-making, bridging advanced analytics with real-time clinical practice.