基于脑电信号Hilbert-Huang变换的睡眠分期研究  

Automatic Sleep Stage Classification Based on Hilbert-Huang Transform Method of EEG.

作　　者：李谷[1] 范影乐[1] 李轶[1] 庞全[1] 

机构地区：[1]杭州电子科技大学生物医学工程及仪器研究所,浙江杭州310018

出　　处：《航天医学与医学工程》2007年第6期458-463,共6页Space Medicine & Medical Engineering

基　　金：国家自然科学基金资助项目(60302027);浙江省教育厅科研计划项目(20030620)

摘　　要：目的研究基于脑电信号Hilbert-Huang变换的睡眠自动分期方法。方法对睡眠脑电信号进行Hilbert-Huang变换,求出具有物理意义的瞬时频率,并得到脑电信号在频率上的能量分布,作为睡眠脑电信号各个时期的特征,最终利用最近邻模式分类方法对睡眠各阶段进行分期决策。结果通过对560个睡眠脑电信号样本进行分期,平均正确率达到81.7%。结论经Hilbert-Huang变换得到的睡眠脑电信号特征,可以作为睡眠分期的有效分类依据。Objective To present a new application of Hilbert-Huang transform method for automatic sleep stage classification. Methods Instantaneous frequencies of sleep EEG with physical meaning and energy-frequency distribution used as feature parameters for each stage were computed with the Hilbert-Huang transform. Finally, pattern recognition method of nearest neighbors was applied to optimal classification. Results Five hundred and sixty samples were picked from sleep EEG , and classification was made. The mean rate of accuracy was as high as 81.7%. Conclusion Automatic sleep stage classification can be made effectively from features of sleep EEG by Hilbert-Huang transform.

关 键 词：HILBERT-HUANG变换 脑电信号 睡眠分期 固有模态函数 

分 类 号：R318.04[医药卫生—生物医学工程]