摘要
癫痫是一种常见的以反复癫痫性发作为特征的慢性神经系统疾病。癫痫性发作的自动检测是通过机器学习及数据挖掘等方法对癫痫发作脑电自动识别的一种技术。如何设计合适的脑电特征提取方法是有效完成癫痫性发作自动检测的关键所在。文中系统总结了用于癫痫性发作自动检测的脑电特征提取方法,分别从时域分析、频域分析、时频分析、非线性动力学、图论、癫痫计算模型6个方面将已有的癫痫脑电特征提取方法进行归类,并对每类方法的基本原理和设计思想进行了系统的阐述。
Epilepsy is a serious brain disorder characterized by recurrent and transient epileptic seizures. Automated epileptic seizure detection is the technique for recognizing epileptic EEGs automatically through the machine learning and data mining methods. How to design an appropriate feature extraction method plays a key role in realizing the seizure detection successfully. This paper systemically summarizes the fundamental theories and designing procedures of existing EEG feature extraction methods from the perspective of time-do- main analysis, frequency-domain analysis, time-frequency analysis, non-linear dynamics, graph-domain analysis and computational models.
出处
《西北大学学报(自然科学版)》
CAS
CSCD
北大核心
2016年第6期781-788,794,共9页
Journal of Northwest University(Natural Science Edition)
基金
国家自然科学基金资助项目(61473223)
陕西省自然科学基础研究计划基金资助项目(2014JM1016)
关键词
癫痫
癫痫性发作
脑电
自动检测
特征提取
epilepsy
epileptic seizure
electroencephalogram (EEG)
automated seizure detection
feature extraction