摘要
脑电中癫痫波的自动检测与分类是临床上很有意义的工作。本文采用一种‘分层次、多方法整合’途径，把自适应预测、小波变换、人工神经网络、模糊识别、专家系统等信号处理技术有机地结合在处理的不同层次（如初筛、分解、特征提取、识别与分类等）上，建立起一种新的分析框架。
utomatic detection and classification of epileptic waves in EEG is of great significance clinically. A `hierarchical multimethod approach' was used, which integrated various signal processing techniques (e.g.: adaptive prediction, wavelet transform, artificial neural network, fuzzy recognition and expert system) at various levels of the processing process(e.g.: preliminary screening, basic waveform decomposition, feature extraction, recognition and classification), to establish a new analysis system. Initial clinical results obtained were encourging.
出处
《中国生物医学工程学报》
CAS
CSCD
北大核心
1998年第1期1-11,共11页
Chinese Journal of Biomedical Engineering
基金
自然科学基金
关键词
癫痫波
连续小波变换
模糊神经网络
专家系统
Epileptic wave
Continuous wavelet transform
Fuzzy neural network
Expert system
