信号长度对基于心率校正心率变异性分析准确性的影响  

Effect of signal length during electrocardiography on accuracy of heart rate variability based on heart rate correction

作　　者：鄢苏鹏 魏良[1] 龚渝顺[1] 胡厚源 李永勤[1] YAN Supeng;WEI Liang;GONG Yushun;HU Houyuan;LI Yongqin(Faculty of Biomedical Engineering and Imaging Medicine,Army Medical University(Third Military Medical University),Chongqing,400038,China;Department of Cardiology,Frist Affiliated Hospital,Army Medical University(Third Military Medical University),Chongqing,400038,China)

机构地区：[1]陆军军医大学(第三军医大学)生物医学工程与影像医学系,重庆400038 [2]陆军军医大学(第三军医大学)第一附属医院心血管内科,重庆400038

出　　处：《陆军军医大学学报》2023年第17期1860-1868,共9页Journal of Army Medical University

基　　金：国家自然科学基金面上项目(62271490)。

摘　　要：目的 观察信号长度对基于心率校正心率变异性(heart rate variability, HRV)分析结果的影响,探索获得准确HRV分析所需采集信号的最短时间长度。方法 采集2017年5月至2019年11月在陆军军医大学第一附属医院体检或就诊的116名正常对照者和72名肥厚型心肌病(hypertrophic cardiomyopathy, HCM)患者的24 h心电数据,运用基于心率校正的HRV方法计算临床常用的9个指标。以24 h结果为标准,对比3种信号长度(5、30、60 min)计算结果的准确性和可靠性。结果 用基于心率校正的HRV方法对3种长度的信号进行分析,HCM患者的9个HRV指标均显著低于正常对照者(P<0.05)。当信号长度为5、30和60 min时,随着信号长度的增加,HRV结果和24 h结果的相关系数逐渐增大,分别为0.36±0.22、0.76±0.14和0.97±0.04 (P<0.05),和24 h结果的相对误差逐渐减小,分别为(28.09±22.96)%、(7.78±4.78)%和(3.27±1.79)%(P<0.05),受试者工作特征曲线下面积逐渐增大,分别为0.75±0.04、0.87±0.08和0.89±0.08 (P<0.05)。结论 短时HRV分析对HCM患者风险评估结果的准确性随信号长度的减小而降低。当使用基于心率校正的HRV分析方法时,获得准确结果所需采集信号的最短时间为60 min。Objective To investigate the effect of different signal length during electrocardiography(ECG)on results of heart rate(HR)-corrected heart rate variability(HRV)analysis and explore the minimum time window of signal collection for accurate HRV analysis.Methods 24-hour electrocardiogram signal were collected from 116 normal controls and 72 hypertrophic cardiomyopathy(HCM)patients admitted in the First Affiliated Hospital of Army Medical University during May 2017 to November 2019.Nine commonly used HRV metrics in clinical practice were obtained by HR-corrected HRV analysis.The HRV results of 24-hour signal length served as reference standards,and then the accuracy and reliability of HRV results at a signal length of 5,30 and 60 min were compared,respectively.Results In results of HR-corrected HRV analysis at 3 different signal lengths,all the 9 HRV metrics were significantly lower in the HCM patients than the normal controls(P0.05).With the increment of the signal length,the correlation coefficient between short-term HRV results and 24 h HRV results was elevated gradually(P0.05),as 0.36±0.22,0.76±0.14,and 0.97±0.04 for the length of 5,30 and 60 min,respectively,and the relative error between short-term HRV results and 24 h HRV results was reduced(P0.05),as(28.09±22.96)%,(7.78±4.78)%,and(3.27±1.79)%for the 3 signal lengths,with an AUC value of 0.75±0.04,0.87±0.08,and 0.89±0.08,respectively(P0.05).Conclusion The accuracy of short-term HRV analysis is decreased with the shortening of signal length in assessment for risk of HCM patients.For HR-corrected HRV analysis,the minimum time window of signal collection is 60 min in order to obtain accurate HRV results.

关 键 词：信号长度 心率校正 心率变异性 准确性 

分 类 号：R312[医药卫生—基础医学] R540.41R542.2