A Machine Learning Approach to Seizure Detection in a Rat Model of Post-Traumatic Epilepsy

Abstract

Abstract Epilepsy is a common neurologic condition frequently investigated with rodent models, with seizures identified by electroencephalography (EEG). Given technological advances, large datasets of EEG amenable to machine learning approaches for identification of seizures are widespread. While such approaches have been explored for human EEGs, machine learning approaches to identifying seizures in rodent EEG are limited. We utilized a predesigned deep convolutional neural network (DCNN), GoogLeNet, to classify images for seizure recognition. Training images were generated through multiplexing spectral content (scalograms), kurtosis, and entropy for two-second EEG segments. Over 2200 hours of EEG data were scored for the presence of seizures, with 95.6% of seizures identified by the DCNN and a false positive rate of 34.2% (1.52/hr), as compared to visual scoring. Multiplexed images were superior to scalograms alone and a DCNN trained specifically for the individual animal was superior to using DCNNs across animals. For this dataset the DCNN approach is superior to an algorithm utilizing total variation following wavelet decomposition. We demonstrate the novel use of a predesigned DCNN constructed to classify images, utilizing multiplexed images of EEG spectral content, kurtosis, and entropy, to rapidly and objectively identifies seizures in a large dataset of rat EEG with high sensitivity.