Absence epilepsy is a neurological condition characterized by abnormally synchronous electricalactivity within two mutually connected brain regions, the thalamus and cortex, that results in seizuresand affects more than 6.5 million people. Epilepsy is commonly studied through the use of theelectroencephalogram (EEG), a device that monitors brain waves over time. In this study, weintroduced machine learning models to predict epileptic seizures in two ways, one to train logisticregression models to provide an accurate decision boundary to predict based off frequency features,and second to train convolutional neural networks to predict based off spectral power images fromEEG. This pipeline employed a two model approach, using logistic regression and convolutionalneural networks to predict seizures. The evaluation, performed on data from 9 mice, achievedprediction accuracies of 98%. The proposed methodology introduces a novel aspect of lookingat predicting absence seizures, which are known to be short events, in addition to the comparisonbetween a time-dependent and time-agnostic seizure prediction classifier. The overall goal of theseexperiments were to build a model that can accurately predict whether or not a seizure will occur.
bioRxiv Subject Collection: Neuroscience