Epilepsy Detection Method Based on the Time-gated Feature Network

Abstract Epilepsy is a nervous system disease, which caused by abnormal discharge of brain neurons. The clinical manifestations are generalized seizures, clonus, loss consciousness, and shock. An electroencephalogram (EEG) can accurately capture the changes in EEG activities. Therefore, signals used to detect seizures. In this paper, an epilepsy detection model based on time-gated feature network (TFGN) proposed. Firstly, original signal preprocessed, preprocessed sent into TFGN integrates extraction, selection, classification obtain results epilepsy. Through verification data from different ages channels, accuracy higher than that traditional model, validity comprehensiveness verified.