Development of a Closed Loop Stimulation System for Epilepsy Therapy

Abstract

University of Freiburg’s Cluster of Excellence BrainLinks-BrainTools aims to develop smart devices to interface with the brain. Among these, closed-loop stimulation devices promise new chances for patients suffering from pharmacoresistant epilepsy, assuming that effective stimulation parameters can be found. As the mechanisms of electrical brain stimulations are complex and poorly understood, animal experiments are needed to obtain insight into optimal parameter settings. Besides these, the loop can only be closed by reliable seizure detection and prediction algorithms with high performance and at the same time low computational complexity. We present our results in this respect, based on the widely accepted kainate model, thought to resemble human temporal lobe epilepsy. More than 400 seizures could be recorded in more than 2500 h of synchronized video-EEG-acquisition. Video- and EEG-based detection algorithms were programmed and seizure prediction methods tested. Those with the best performance in regard to reliability, celerity and power consumption will be used in a closed-loop setup. This is a first step to evaluate closed-loop stimulation approaches and therewith widening possibilities for future clinical applications.