Seizure Prediction in Epilepsy: Does a Combination of MethodsHelp?

摘要

Epilepsy, as one of the most common neurological diseases, affects up to 1 % ofthe world population. Due to the unforseeable nature of epileptic seizures, manypatients suffer from severe impairments (see Chapter 1). In addition to establishedtherapies like antiepileptic medication or surgical resections, closed-loop predictionand intervention methods could open up a new window of treatment strategies. Forthis prospect, several methods have been proposed for the prediction of epilepticseizures (see Chapter 2). By means of linear and nonlinear time series analysis,predictive features have been found in several studies (for a review see [1]). I f preictalchanges in the EEG could be detected early and with a sufficient sensitivity andspecificity, alarms could be raised to warn the patients. Furthermore, automaticallytriggered interventions like focal drug applications or electrical stimulations wouldbe feasable [2, 3].Bjorn Schelter