A connectionist perspective on detection and control of epileptic seizures

Epileptic seizures correspond to episodes of increased rhythmicity of the normally chaotic electrical activity in biological neural networks (BNNs). A connectionist perspective is presented whereby artificial neural networks (ANNs) are used to learn the chaotic dynamics of the biological neural networks. The ANNs are used to detect a change to a rhythmic mode in the BNNs, then employ nonlinear dynamics to restore the BNNs to their chaotic mode.