基于深度学习的癫痫脑电信号分析与预测

Analysis and Prediction of Epilepsy EEG Signals Based on Deep Learning

癫痫(Epilepsy)是关于大脑功能障碍的慢性神经系统疾病之一,目前研究表明其病理原因是大脑的神经元发生了异常突发放电,准确诊断该疾病需要长时间的脑电监测,而人工识别工作量巨大且具有主观性。深度学习是一种构造多层神经网络的机器学习方法,具有发现数据中隐藏的分布式特征表示的能力。针对癫痫患者的脑电信号,本文介绍癫痫脑电信号的特征提取、脑电信号的分析及脑电信号的分类方法。为采用深度学习对癫痫进行预测提供理论依据。

Epilepsy is one of the chronic nervous system diseases related to brain dysfunction. The current research shows that the pathological cause is abnormal discharge of neurons in the brain. The accurate diagnosis of the disease requires long-term EEG monitoring, but manual identification is a huge workload and has subjective intention. Deep learning is a machine learning method for constructing multi-layer neural networks, having the ability to discover hidden feature representations in data. In view of the EEG signals of patients with epilepsy, this paper introduces the feature extraction of EEG signals, the analysis of EEG signals and the classification of EEG signals, hoping to provide a theoretical basis for predicting epilepsy by deep learning.