基于心率变异性分析的睡眠分期方法研究

发布时间：2018-04-15 22:29

本文选题：心率变异性 + 主成分分析　；参考：《生物医学工程学杂志》2016年03期

【摘要】：为实现对不同阶段睡眠的快速、便捷分期,本文实验分析了心率变异性(HRV)与睡眠分期的相关性,利用支持向量机(SVM)实现了基于HRV信息的睡眠自动分期的算法。对天津市胸科医院的33例临床心电数据进行了R-R提取和主成分分析(PCA),并利用SVM对睡眠中的不同阶段进行建模和预测,将分期的预测结果与基于脑电金标准的睡眠分期标注结果比对,对于三期睡眠的预测准确度超过80%,说明HRV与睡眠各期具有良好的相关性。该方法是对传统睡眠分期方法的一种补充,具有实际使用价值。
[Abstract]:In order to realize the rapid and convenient stage of sleep in different stages, the correlation between HRV (heart rate variability) and sleep stage is analyzed experimentally, and the algorithm of automatic sleep staging based on HRV information is realized by using support vector machine (SVM).R-R extraction and principal component analysis (PCA) were performed on 33 cases of clinical ECG data from Tianjin chest Hospital. SVM was used to model and predict the different stages of sleep.Comparing the prediction results of stages with the results of sleep staging tagging based on EEG gold standard, the prediction accuracy of three-phase sleep is over 80, which indicates that HRV has a good correlation with sleep stages.This method is a supplement to the traditional sleep staging method and has practical value.
【作者单位】：天津工业大学电子与信息工程学院;天津市医学电子诊疗技术工程中心;天津市胸科医院;
【基金】：天津市应用基础与前沿技术研究计划资助项目(13JCYBJC37800) 教育部留学归国科研启动基金(第49批)
【分类号】：R740

【相似文献】