Authors :
Presenting Author: Anuraag Velamati, BS – University of California, Davis
Rocelle Evangelista, BS – University of California, Davis; Sheela Toprani, MD, PhD – University of California, Davis
Rationale: There have been multiple approaches in the past to treat seizures by understanding how seizures traverse in an individual's brain. Lately, the emphasis has shifted from traditional approaches to understanding seizure networks. Current treatments, though available, underutilize this network. Our goal is to improve surgical treatment design by identifying key nodes within the seizure network. Low-frequency electrical brain stimulation is a known methodology for estimating baseline network functional connectivity. This approach leverages responses to electrical stimulation, referred to as cortico-cortical evoked potentials (CCEPS), to discern the interconnected regions within a network. Using CCEPS, we can gain insight into the network's interactions, facilitating the evaluation of network dynamics and relationships. We present a case study of a patient undergoing phase II monitoring, where the network's effective connectivity was investigated through analysis of the CCEPs responses.
Methods: We employed low-frequency electrical brain stimulation to study seizure propagation in a 23 year old right handed caucasian male with refractory epilepsy undergoing phase II monitoring for surgical evaluation. CCEPs results were analyzed by computing average and SEM of channel responses, saving waveform data for each segment. Utilizing graph theory, we assessed out-degree (influence) and in-degree (influenceability). Detecting N1 and N2 peaks, we characterized their presence in the hypothesized network, revealing insights into neural response speed and strength. We then rank the electrodes, based on their in-degree and out-degree, to assess the influence and influenceability of the seizure focus, on the overall network.
Results: The analysis of CCEPs responses uncovered significant patterns within the brain network. When examining in-degree connectivity, we observed that 11 out of the top 15 electrodes exhibiting high in-degree connections were located in white matter. Likewise, in out-degree connectivity, a similar trend emerged, with 8 out of the top 15 electrodes displaying high out-degree connections also belonging to white matter.
Conclusions: We observed certain patterns of seizure propagation in the brain's network. For this patient, the electrode locations in the white matter had higher influence and influenceability. A similar pattern was observed even with the other patients in our analysis. Though we cannot conclude, due to the need for more data, our current checkpoint gave us a new insight into the seizure network dynamics. Our analysis has unveiled that electrodes situated within the white matter exert a higher degree of influence on the global neural network dynamics, irrespective of their presence in the seizure network. This observation underscores the significant role played by white matter in shaping the overall network dynamics during epileptic events. The next step is to investigate the changes in the network connectivity by taking only the gray matter electrodes into consideration and look whether the electrodes with higher influenceability belong to the seizure onset zone.
Funding: N/A