论文部分内容阅读
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。
Sleep stage is to study the basis of sleep and related diseases, is to complete the premise of quality assessment of sleep, has important clinical significance. In recent years, automatic staging of sleep-based brain electrical signals based on computer technology has become a research hotspot, and some achievements have been made. This paper introduces the basic knowledge of sleep staging and electroencephalography (EEG), and discusses in detail the two key techniques of automatic sleep staging based on EEG (feature extraction and pattern recognition), and compares wavelet transform and Hilbert-Huang transform The advantages and disadvantages of two commonly used methods of EEG feature extraction, artificial neural network and support vector machine pattern recognition and their application in sleep stage are summarized, and the research status and development trend in this area are summarized.