基于短时心电图时间间期序列和血压序列的耦合性分析

发布时间：2018-03-08 17:28

本文选题：心电图时间间期　切入点：血压(BP)　出处：《北京协和医学院》2017年硕士论文　论文类型：学位论文

【摘要】：背景:自主神经系统(ANS)在心血管的生理调节方面扮演着重要角色。ANS失衡可能使心血管功能受到损伤,导致多种心血管疾病的发生、发展。心电图间期信号和动脉血压信号受自主神经的调节,呈现出一定的动力学特性,如心率变异性(HRV)和血压变异性(BPV)。这些心血管生理信号的动力学相互作用包含着自主神经调节的重要信息。本研究拟以连续心电图间期序列和收缩压序列(SBP)为基础,分别进行RR间期(RRI)与SBP以及RRI与QT间期(QTI)的短时联合分析,通过表征RRI-SBP以及RRI-QTI的因果相互作用,探讨自主神经状态的无创评估方法。方法:选用PhysioNet数据库的PRCP子数据库(n=10)作为RRI-SBP联合分析的数据源。从动态和稳态两个视角,研究直立倾斜(HUT)引起体位改变前后,以及不同速度改变体位之后,RRI与SBP间耦合性的变化。慢速体位变化(ST)和快速体位(RT)变化分别为在50和2 s之间从水平仰卧升至75°倾斜。提取逐拍RRI和SBP数据后,运用交叉时频分析和信息分解方法,结合时域和短时分形指数(α1),进行RRI和SBP时间序列的联合分析。选用THEW数据库中的Normal子数据库和ESRD子数据分别作为RRI-QTI联合分析的对照组(n=189)和病理实验组(n=43)数据源。采用基于自回归(AR)模型的短时因果交叉谱方法,通过分析从RRI到QTI(RRI→QTI)方向的线性耦合,探讨在生理和病理条件下,基于昼夜节律的自主神经行为刻画方法。结果:针对RRI-SBP分析,信息分解分析结果表明,所有的显著差异集中在压力反射导致心率变化de后向反馈回路(SBP→RRI),ST后心率的可预测性较平卧时显著增高(0.416±0.067vs0.626±0.127),压力反射支路的SBP-RRI耦合性升高。而在RRI→SBP方向,HUT对其几乎没有影响。ST和RT之前,所有的同类指标相比均无显著差异。ST和RT之后的稳态,虽然RRI无显著差异,但较之ST之后,RT之后RRI 的变异系数显著升高(0.054±0.014 vs 0.074±0.027),α1 显著降低(1.45±0.25 vs 1.28±0.27)。同时,交叉时频分析结果揭示了 ST和RT过程中自主神经不同的动态反应行为。研究证明了信息分解方法的有效性,可明确区分心率与血压相互作用时的前向反馈和后向反馈的主导因果方向,而且可反映HUT前后信号可预测性的变化。针对RRI-QTI分析,因果交叉谱结果表明,Normal组的低频和高频段的传递函数增益(GLF、GHF)以及LF/HF的昼夜结果之间存在显著性差异,而相应指标在ESRD组中均无昼夜差异;Normal组的昼夜指标之比GLF-d/GLF-n以及(LF/HF)d/(LF/HF)n指标均显著大于ESRD组的结果,而GHF-d/GHF-n则无统计学差异。研究结果提示,相较于正常人(Normal组),ESRD组出现昼夜节律的消失意味着夜间交感神经异常活跃的病理现象。结论:本课题从RRI-SBP以及RRI-QTI的短时联合分析两个方面,结合时域和频域分析以及线性和非线性分析,探索基于动态心电图间期时间序列和逐拍血压序列的心脏自主神经行为的评估方法,为临床应用心脏自主神经状态评估心脏安全提供可靠依据。
[Abstract]:Background: the autonomic nervous system (ANS) plays an important role in regulation of.ANS may cause the cardiovascular function damage in cardiovascular physiology, cause a variety of cardiovascular disease development. The regulation of blood pressure and ECG signal interval signal artery is regulated by the autonomic nervous system, showing the dynamic characteristics, such as heart rate variability (HRV) and blood pressure variability (BPV). The dynamic interaction of these cardiovascular physiological signal contains important information in autonomic regulation. This study intends to continuous ECG interval and systolic blood pressure (SBP) sequence based, respectively RR interval (RRI) and SBP, RRI and QT interval (QTI) combined with the analysis of the short-term. Characterized by RRI-SBP and RRI-QTI causal interactions, noninvasive evaluation methods of autonomic nervous state. Methods: the PhysioNet database PRCP database (n=10) as RRI-SBP joint Analysis of the data source. From the two aspects of dynamic and static, tilt (HUT) cause postural changes, and different speed changing positions, coupling changes between SBP and RRI. The slow postural changes (ST) and fast position (RT) changes were between 50 and 2 s from the level of supine rose to 75 degrees tilt. The extraction of beat to beat RRI and SBP data, using frequency analysis and cross information decomposition method, combining time-domain and short-time fractal index (alpha 1), combined with analysis of RRI and SBP time series. The THEW database Normal database and ESRD data respectively as the control group combined analysis of RRI-QTI the experimental group (n=189) and pathology (n=43) data source. Based on the autoregressive (AR) short-term causal model of cross spectrum method, through analysis from RRI to QTI (RRI, QTI) linear coupling direction, in physiological and pathological conditions, circadian rhythm based on self The main method to characterize the behavior of nerve. Results: according to RRI-SBP analysis, information decomposition analysis results show that the concentration difference of all lead to changes in the de heart rate baroreflex to feedback loop (SBP, RRI), ST heart rate predictability is recumbent significantly higher (0.416 + 0.067vs0.626 + 0.127), SBP-RRI coupling pressure the reflection branch in RRI and SBP increased. The direction of HUT have little effect on the.ST and RT before the steady state similar indicators all showed no significant difference between.ST and RT, while RRI had no significant difference, but compared to ST, after RT, the coefficient of variation of RRI was significantly increased (0.054 + 0.014 vs 0.074 + 0.027), alpha 1 was significantly decreased (1.45 + 0.25 vs 1.28 + 0.27). At the same time, cross time-frequency analysis results reveal the dynamic response behavior of autonomic nerve of different ST and RT process. The research proves the validity of information decomposition method, can be clearly distinguished Feed forward and backward direction leading causal feedback of heart rate and blood pressure during the interaction, but also can reflect the HUT signal before and after changes in predictability. According to RRI-QTI analysis, causal cross spectrum results show that the transfer function gain Normal group of low and high frequencies (GLF, GHF) there are significant differences between LF/HF and the circadian results in the ESRD group and the corresponding index showed no difference between day and night; day and night index in the Normal group the ratio of GLF-d/GLF-n (LF/HF) and d/ (LF/HF) n index were significantly greater than group ESRD results, but there were no significant differences in GHF-d/GHF-n. The results of the study suggest that, compared to the normal group (Normal group), ESRD the emergence of circadian rhythm disappearance means pathological phenomenon of abnormal sympathetic activity at night. Conclusion: the combined analysis from two aspects of RRI-SBP and RRI-QTI combined with short time domain and frequency domain analysis and linear and nonlinear analysis To explore the evaluation method of cardiac autonomic behavior based on dynamic electrocardiogram interval time series and beat to beat blood pressure sequence, so as to provide reliable evidence for clinical application of cardiac autonomic nervous system to evaluate cardiac safety.

【学位授予单位】：北京协和医学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R540.41

【参考文献】