• A Journal of Neural Engineering study presents a spatio-temporal transformer—“the seizure embedding map”—for comparing patients by ictal intracranial EEG (iEEG) features at scale.1
  • Invasive treatment planning for medication-resistant epilepsy depends on interpreting iEEG to identify seizure onset patterns and locations.1
  • Matching a new patient’s seizures to prior cases remains subjective because implant strategies, electrode placements, and seizure onset zones differ across individuals and centers.1
  • The transformer embeds ictal iEEG so new patients can be compared with prior cases despite heterogeneous implants.1
  • The approach targets epilepsy planning and scalable analysis of intracranial signals across varied recording setups.1 1

Footnotes

  1. https://iopscience.iop.org/article/10.1088/1741-2552/ae805d 2 3 4 5 6