{
  "abstract": "Aperiodic electroencephalographic activity (AA), represented by the EEG frequency spectrum once rhythms are excluded, can be parameterised by the slope of the power spectrum. The aperiodic exponent (AE) is one method of parameterising this slope and has been shown to correlate with the ratio of inhibitory to excitatory activity.This study explored changes in AA from an EEG dataset of 2840 seizures among 248 patients. AEs were computed for fifteen-second windows of preictal and postictal data and expert-labelled interictal and ictal data.Peri-ictal AE differed significantly between leads that were or were not involved in the seizure. AE increased globally from interictal to preictal state, starting at least 30 seconds before ictus. AE decreased in the 30 seconds preceding ictus, specifically in the seizure-involved leads.If seizures arise from cortex with a greater excitatory/inhibitory ratio, a decreasing preictal AE aligns with prior research suggesting that the AE is inversely proportional to this ratio. The mixture of global, and lead-specific changes may relate to both local, and network dynamics. The early global changes emerging preictally may offer insight into network dynamics that modulate seizures, and offer a quantitative predictive clinical tool.rohan.kandasamy@nhs.net",
  "authors": [
    {
      "affiliations": [
        "UCL Institute of Neurology",
        "UCLH Department of Clinical Neurophysiology"
      ],
      "name": "Kandasamy Rohan"
    },
    {
      "affiliations": [
        "Department of Engineering, University of the West of UK"
      ],
      "name": "Western David"
    },
    {
      "affiliations": [
        "North Bristol Trust"
      ],
      "name": "May Felix"
    },
    {
      "affiliations": [
        "North Bristol Trust"
      ],
      "name": "Canham Luke"
    }
  ],
  "title": "134 Preictal aperiodic activity in EEG shows changes specific to seizure regions",
  "uid": "daeb3a5d-f82e-5ca3-944c-443d2c77acbe"
}