模糊区间信息熵用于充血性心衰患者的检测

目的传统心率变异性图形化指标（直方图和散点图）检测区分充血性心衰患者和健康人时难以量化，针对这一问题，本文提出一种新的心率变异性图形化分析方法——RR间期序列归一化直方图及其量化指标，用于检测区分充血性心衰患者和健康人。方法首先定义RR间期序列归一化直方图，在此基础上定义基于模糊理论的量化指标——模糊区间信息熵FuzzyRIEn，最后利用120例临床试验数据（60例充血性心衰患者和60例健康人）分析对比指标FuzzyRIEn和前期研究提出的3个量化指标：中心一边缘比（center—edgeratio，CER）、累积能量（cumulativeenergy，CE）和区间信息熵（range information entropy，RIEn）在两组问的统计学差异。结果对比结果显示，CER（P=0．418），CE（P=0．262）和RIEn（P=0．068）在两组间均无显著统计学差异，而FuzzyRIEn（P=0．023）在两组间统计学差异显著。结论利用模糊区间信息熵FuzzyRIEn检测区分充血性心衰有较高的临床诊断价值。

Fuzzy range information entropy to detect congestive heart failure

Objective Histogram and scatter plot are two important graphical indices for heart rate variability （HRV） analysis which are difficult to quantify the differences between the patients with congestive heart failure and healthy control subjects. This study proposes a new graphical method for HRV analysis, named RR sequence normalized histogram. Methods Firstly, RR sequence normalized histogram was defined. Meanwhile, the index of fuzzy range information entropy （FuzzyRIEn） based on the fuzzy sets was defined. Te verify the validity of FuzzyRIEn, a total of 120 subjects （60 patients with congestive heart failure and 60 healthy subjects） were enrolled and the comparison between FuzzyRIEn and the other three indices [ center- edge ratio （CER）, cumulative energy （CE） and range information entropy （RIEn）J that we proposed earlier were analyzed. Results Wilcoxon rank sum test showed that there were no statistical differences in CER （P = 0.418）, CE （P = 0.262） and RIEn （P = 0.068） between the two groups meanwhile the difference in FuzzyRIEn （P = 0. 023） was significant. Conclusions FuzzyRIEn might be valid for detecting congestive heart failure in clinical applications.