XGboost-based Method for Seizure Detection in Mouse Models of Epilepsy

摘要

Epilepsy is a chronic neurological disease which affects over 50 million people worldwide, caused by the disruption of the finely tuned inhibitory and excitatory balance in brain networks, manifesting clinically as seizures. Electroencephalographic (EEG) monitoring in rodent disease models of epilepsy is critical in the understanding of disease mechanisms and the development of anti-seizure drugs. However, the visual annotation of EEG traces is time-consuming, and is complicated by different models and seizure types. Automated annotation systems can help to solve these problems by reducing expert annotation time and increasing the throughput and reliability of seizure quantification. As machine learning is becoming increasingly popular for modelling sequential signals such as EEG, several researchers have tried machine learning to detect seizures in EEG traces from mouse models of epilepsy. Most existing work can only detect seizures in single mouse models of epilepsy and research on multiple mouse models has been limited to-date. In this work, we developed a Teager-Kaiser energy operator (TKEO)-based method that mimics how experts detect seizures in an intra-amygdala kainic acid (IAKA) mouse model of epilepsy. Furthermore, we propose a machine learning-based method which can review large volumes of data and discover specific trends and patterns that may not be apparent to humans. We compared the performance of the TKEO-based and machine learning-based method on the IAKA mouse model. We further tested these two methods on a Dravet syndrome (DS) mouse model of epilepsy to see whether these two methods could generalize to detect seizures in another mouse model of epilepsy.