李向欢 杜玉晓 凌宇
（广东工业大学自动化学院，广东 广州 510006）
摘要：癫痫是大脑神经元突发性异常放电，导致短暂的大脑功能障碍的一种慢性疾病。癫痫脑电信号的高频振荡是一种可靠的癫痫发生生物标志物。为了有效地诊断和治疗癫痫，对癫痫脑电高频振荡进行准确地检测至关重要。癫痫脑电高频振荡自动检测算法先从脑电信号中提取相关特征，再利用机器学习算法来识别分类高频振荡。首先，介绍癫痫脑电高频振荡的定义以及临床研究意义；然后，对癫痫脑电高频振荡的特征提取和特征分类进行总结；最后，阐述癫痫脑电高频振荡自动检测的研究现状和未来的研究方向。
关键词：癫痫；脑电图；高频振荡；特征提取；特征分类
中图分类号：R742.1 文献标志码：A文章编号：1674-2605(2023)03-0001-09
DOI：10.3969/j.issn.1674-2605.2023.03.001
Research Progress in Automatic Detection of Epileptic EEG High-frequency Oscillations
LI Xianghuan DU Yuxiao LING Yu
(School of Automation, Guangdong University of Technology, Guangzhou 510006, China)
Abstract:Epilepsy is a chronic disease caused by sudden abnormal discharge of brain neurons, which leads to temporary brain dysfunction. The high-frequency oscillation of epileptic EEG signals is a reliable biomarker for epilepsy occurrence. In order to effectively diagnose and treat epilepsy, accurate detection of high-frequency oscillations in epileptic EEG is crucial. The automatic detection algorithm for high-frequency oscillations in epilepsy EEG first extracts relevant features from EEG signals, and then uses machine learning algorithms to identify and classify high-frequency oscillations. Firstly, introduce the definition and clinical research significance of high-frequency oscillations in epileptic EEG; then, summarize the feature extraction and classification of high-frequency oscillations in epileptic EEG; finally, the current research status and future research directions of automatic detection of high-frequency oscillations in epilepsy EEG are elaborated.
Keywords: epilepsy; EEG; high-frequency oscillations; feature extraction; feature classification
