基于脑电信号的自动睡眠分期研究进展

Research Progress of Automatic Sleep Staging Based on Electroencephalogram Signals

睡眠分期是研究睡眠及相关疾病的基础,是完成睡眠质量评估的前提,具有重要临床意义。近年来,基于计算机技术的睡眠脑电信号自动分期成为研究热点,并取得了一些成果。本文介绍了睡眠分期与脑电信号的基础知识,详细论述了基于脑电信号的自动睡眠分期研究中的两个关键技术——特征提取和模式识别,比较了小波变换、Hilbert-Huang变换两种常用的脑电特征提取方法,和人工神经网络、支持向量机两类模式识别方法的优缺点及其在睡眠分期中的应用,总结了近几年该领域的研究现状和发展趋势。

The research of sleep staging is not only a basis of diagnosing sleep related diseases but also the precondi- tion of evaluating sleep quality, and has important clinical significance. In recent years, the research of automatic sleep staging based on computer has become a hot spot and got some achievements. The basic knowledge of sleep staging and electroencephalogram （EEG） is introduced in this paper. Then, feature extraction and pattern recogni- tion, two key technologies for automatic sleep staging, are discussed in detail. Wavelet transform and Hilbert-Huang transform, two methods for feature extraction, are compared. Artificial neural network and support vector machine （SVM）, two methods for pattern recognition are discussed. In the end, the research status of this field is summa- rized, and development trends of next phase are pointed out.