Cellular neural networks for the anticipation of epileptic seizures
Ronald Tetzlaff, D. Weiss · 2003
In this paper, a new feature extraction method for the analysis of brain electrical activity in epilepsy is introduced. We propose a new prediction algorithm, which is based on discrete time-delay type Cellular Neural Networks (DTCNN) with nonlinear weight functions. In the results distinct changes of calculated features can be observed some minutes before the onset of an epileptic seizure. A detailed discussion will be given.