[1]赵天夏,王新安,李秋平,等.基于心率变异率特征值的心律失常评估研究[J].计算机技术与发展,2023,33(01):21-26.[doi:10. 3969 / j. issn. 1673-629X. 2023. 01. 004]
　ZHAO Tian-xia,WANG Xin-an,LI Qiu-ping,et al.Study of Algorithms and Analytical Evaluation Based on Eigenvalues of ECG Signals[J].,2023,33(01):21-26.[doi:10. 3969 / j. issn. 1673-629X. 2023. 01. 004]
点击复制

基于心率变异率特征值的心律失常评估研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年01期
页码:
21-26
栏目:
大数据与云计算
出版日期:
2023-01-10

文章信息/Info

Title:
Study of Algorithms and Analytical Evaluation Based on Eigenvalues of ECG Signals
文章编号:
1673-629X(2023)01-0021-06
作者:
赵天夏; 王新安; 李秋平; 邱常沛
北京大学深圳研究生院 集成微系统科学工程与应用重点实验室,广东 深圳 518055
Author(s):
ZHAO Tian-xia; WANG Xin-an; LI Qiu-ping; QIU Chang-pei
Key Laboratory of Integrated Micro-systems,Peking University Shenzhen Graduate School,Shenzhen 518055,China
关键词:
心率变异率; 时域分析; 频域分析; 非线性分析; 特征值提取; 样本熵
Keywords:
heart rate variability; time domain analysis; frequency domain analysis; non-linear analysis; eigenvalue extraction; sample entropy
分类号:
TP31
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 01. 004
摘要:
心率变异率作为一种基于心电信号的疾病分析方法,是临床医学上具有重要参考价值的参数指标。 该文深入研究了心率变异率特征值提取的时域分析方法、频域分析方法和非线性分析方法,针对心律失常的特点,在时域特征值中引入了 pNNx 等心率变异率指标,在非线性特征值中引入了多尺度样本熵。 采用 PKU-IMS 心电数据库的窦性心律数据与MIT-BIH 数据库的心律失常数据,提取了窦性心律组与心律失常组的心率变异率特征值,当选取 95% 的置信区间时,时域分析的特征值 nn50,pnn50,nn100 和 pnn100,频域分析的特征值 vlfp,lfp,hfp 和 lf2hf,以及非线性分析的特征值 τ≥4时,可显著区分窦性心率与心律失常。 由于心率变异率分析也应用于糖尿病、脑血管、呼吸系统等疾病的辅助诊断,因此,该心率变异率特征值分析方法有望推广至相关疾病的评估。
Abstract:
Heart rate variability is a method of disease analysis based on ECG signals and an important reference parameter in clinicalmedicine. The time domain analysis method,frequency domain analysis method and non-linear analysis method of extracting the HRVeigenvalues are thoroughly investigated. According to the characteristics of arrhythmia,the pNNx are introduced in the time domain eigenvalues and the multi-scale sample entropy is introduced in the non-linear eigenvalues. Using sinus rhythm data from the PKU-IMSECG database and arrhythmia data from the MIT - BIH database, the characteristic value of HRV of the sinus rhythm group and thearrhythmia group were extracted. When 95% confidence intervals were selected,the eigenvalues of nn50,pnn50,nn100 and pnn100 inthe time domain analysis,the eigenvalues of vlfp,lfp,hfp and lf2hf in frequency domain analysis,and the eigenvalues of?τ　which greaterthan or equal to 4 in the nonlinear analysis can distinguish sinus heart rate and arrhythmia significantly. Since the analysis of HRV is alsoapplied to the auxiliary diagnosis of diabetes,cerebrovascular and respiratory diseases,the analysis method of characteristic value of HRVstudied is expected to be extended to the evaluation of related diseases.

相似文献/References:

[1]尚宇,张甜. 人工神经网络在HRV分析中的应用研究[J].计算机技术与发展,2017,27(09):141.
　SHANG Yu,ZHANG Tian. Research on Application of Artificial Neural Network in HRV Analysis[J].,2017,27(01):141.

更新日期/Last Update: 2023-01-10