基于符号转移熵和平均能量耗散的睡眠分期分析
发布时间：2018-09-10 16:43
【摘要】：随着生活步骤的加快和工作压力的加大,越来越多的人开始感受到睡眠问题带来的困扰。睡眠质量关系着身体健康与和工作效率,睡眠分期结果是衡量睡眠质量的重要指标和诊治睡眠障碍性疾病的重要途径。EEG是脑部电位活动的反映,通过对睡眠EEG的研究可以获得睡眠时脑部活动概况,所以通过睡眠EEG的分期研究对改善睡眠质量或者诊治睡眠障碍性疾病有很大的意义。 EEG是非线性信号,EEG的非线性分析是目前睡眠研究的热点。ECG也是非线性信号,本文通过基于睡眠EEG、ECG的符号转移熵的睡眠分析方法和基于EEG的熵产生率即平均能量耗散的睡眠分析方法来进行清醒期和非快速眼动睡眠Ⅰ期的分期研究。研究发现:符号转移熵、平均能量耗散很好的体现了睡眠状态的变化,在清醒期较大,在非快速眼动睡眠Ⅰ期较小,并经过差异显著性检验和多样本验证。 经分析认为随着睡眠加深,身体单元不断耦合,因此符号转移熵变小;随着睡眠加深,神经细胞突触连接强度减弱,减弱了基因表达的失衡和无序性趋势,因此熵产生率减小。可见实验结果与理论分析是相符合的。因此符号转移熵、平均能量耗散可以作为睡眠自动化分期参数补充到睡眠分期研究中来,在临床上可以通过多参数分析,达到睡眠分期的更高的准确性。
[Abstract]:As life steps accelerate and work stress increases, more and more people begin to experience sleep problems. Sleep quality is related to physical health and work efficiency. Sleep staging is an important index to measure sleep quality and an important way to diagnose and treat sleep disorders. EEG is a reflection of brain potential activity. A study of sleep EEG provides an overview of brain activity during sleep. Therefore, it is of great significance to improve the quality of sleep or to diagnose and treat sleep disorders by stages of sleep EEG. EEG is the nonlinear analysis of nonlinear signal, which is a hot spot in sleep research. EEG is also a nonlinear signal. In this paper, sleep analysis method based on symbol transfer entropy of sleep EEG,ECG and sleep analysis method of average energy dissipation based on entropy production rate of EEG were used to study the stages of waking and non-REM sleep stages. It was found that the change of sleep state was well reflected in the symbol transfer entropy and the average energy dissipation. It was larger in awake stage and smaller in non-REM sleep stage I, and was verified by difference significance test and multi-sample test. It is concluded that with the deepening of sleep, the body units are coupled continuously, so the symbol transfer entropy becomes smaller; with the deepening of sleep, the synaptic connection intensity of nerve cells weakens, which weakens the imbalance and disordered tendency of gene expression, so the entropy production rate decreases. It can be seen that the experimental results are in agreement with the theoretical analysis. So the symbol transfer entropy and the average energy dissipation can be used as sleep automatic staging parameters to supplement the sleep stage study, and the higher accuracy of sleep staging can be achieved through multi-parameter analysis in clinic.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318
[Abstract]:As life steps accelerate and work stress increases, more and more people begin to experience sleep problems. Sleep quality is related to physical health and work efficiency. Sleep staging is an important index to measure sleep quality and an important way to diagnose and treat sleep disorders. EEG is a reflection of brain potential activity. A study of sleep EEG provides an overview of brain activity during sleep. Therefore, it is of great significance to improve the quality of sleep or to diagnose and treat sleep disorders by stages of sleep EEG. EEG is the nonlinear analysis of nonlinear signal, which is a hot spot in sleep research. EEG is also a nonlinear signal. In this paper, sleep analysis method based on symbol transfer entropy of sleep EEG,ECG and sleep analysis method of average energy dissipation based on entropy production rate of EEG were used to study the stages of waking and non-REM sleep stages. It was found that the change of sleep state was well reflected in the symbol transfer entropy and the average energy dissipation. It was larger in awake stage and smaller in non-REM sleep stage I, and was verified by difference significance test and multi-sample test. It is concluded that with the deepening of sleep, the body units are coupled continuously, so the symbol transfer entropy becomes smaller; with the deepening of sleep, the synaptic connection intensity of nerve cells weakens, which weakens the imbalance and disordered tendency of gene expression, so the entropy production rate decreases. It can be seen that the experimental results are in agreement with the theoretical analysis. So the symbol transfer entropy and the average energy dissipation can be used as sleep automatic staging parameters to supplement the sleep stage study, and the higher accuracy of sleep staging can be achieved through multi-parameter analysis in clinic.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318
【参考文献】
相关期刊论文 前10条
1 吴锋,俞梦孙,张宏金,金璋瑞;心率变化特征与睡眠分期耦合关系研究[J];北京生物医学工程;2003年03期
2 解幸幸;李舒;张春利;李建康;;Lempel-Ziv复杂度在非线性检测中的应用研究[J];复杂系统与复杂性科学;2005年03期
3 刘建平，，郑崇勋;脑电信号的分析──一种探索大脑功能状态及活动规律的途径[J];国外医学.生物医学工程分册;1995年05期
4 李勇;脑电信号现代分析方法的发展[J];国外医学(生物医学工程分册);1996年04期
5 谢松云;张振中;杨金孝;张坤;;脑电信号的若干处理方法研究与评价[J];计算机仿真;2007年02期
6 和卫星;陈晓平;邵s
本文编号：2235022
本文链接：https://www.wllwen.com/yixuelunwen/swyx/2235022.html
