基于短时心率变异特征ST-fApEn的阻塞性睡眠呼吸暂停疾病状态分析方法研究  

Research on the Status Analysis Method of Obstructive Sleep Apnea Based on Sliding Trend Fuzzy Approximate Entropy of Short-term Heart Rate Variability

作　　者：周广敏 刘官正 李一帆 Zhou Guangmin;Liu Guanzheng;Li Yifan(不详;School of Biomedical Engineering and the Key Laboratory of Sensing Technology and Biomedical instruments of Guangdong Province,Sun Yat-sen University,Guangzhou Guangdong 510275,China)

机构地区：[1]中国科学院大学附属肿瘤医院(浙江省肿瘤医院),中国科学院肿瘤与基础医学研究所,浙江杭州310022 [2]中山大学生物医学工程学院,广东省传感技术与生物医疗仪器重点实验室,广东广州510275

出　　处：《航天医学与医学工程》2020年第3期240-245,共6页Space Medicine & Medical Engineering

基　　金：深圳市自由探索项目(201803073000427);2020年广州市高校创新创业(就业)教育-IAB视域下运动健康产业大学生创新创业平台建设项目。

摘　　要：目的提出基于滑动趋势模糊近似熵(ST-fApEn)的短时心率变异性(HRV)分析动态特征,以提高阻塞性睡眠呼吸暂停(OSA)的检测精度及疾病状态的分析能力。方法使用Physionet的ApneaECG数据库中20例正常对照样本和40例OSA样本,分成一系列5min的片段。根据该片段是否发生暂停分为正常对照组(N-N组,20例),OSA患者正常片段组(P-N组,40例)和OSA患者呼吸暂停片段组(P-OSA组,40例)。通过使用经验模态分解(EMD)结合滑动窗的方法来获得RR序列的趋势斜率序列,并对其进行复杂度分析得到ST-fApEn,分析不同疾病状态。结果与SD、RMS、PNN50和LF/HF等HRV静态特征相比,ST-fApEn不仅在3种不同疾病状态组的任意2组都有显著性差异,且OSA的检测精度显著提高到85.0%,敏感性和特异性也分别达到了82.5%和90.0%。结论提出的ST-fApEn动态特征使OSA的检测精度得到显著提高,是鉴别健康人、OSA患者不发病时和发病时不同疾病状态的一个有效检测指标。Objective The dynamic characteristic—sliding trend fuzzy approximate entropy(ST-fApEn)based on short-term heart rate variability(HRV)was put forward to improve the accuracy of obstructive sleep apnea(OSA)and the ability to analyze the disease status.Methods Twenty normal samples and 40 OSA samples from the Apnea-ECG database of Physionet were divided into a series of 5-minute segments.According to whether the segment was suspended or not,the samples were divided into different disease status groups:normal control group(N-N group,20 cases);normal segments of OSA patients(P-N group,40 cases);apnea segments of OSA patients(P-OSA group,40 cases).The ST-fApEn was obtained by analyzing the complexity of the trend slope sequence of RR sequence,got by empirical mode decomposition(EMD)combined with sliding window method to analyze the different disease.Results Compared with the classic HRV static characteristics,such as SD、RMS、PNN50 and LF/HF,ST-fApEn not only showed significant difference in any two groups with three different disease states,but also significantly improved the detection accuracy of OSA to 85.0%,and the sensitivity and specificity reached 82.5%and 90.0%respectively.Conclusion The proposed dynamic characteristic—ST-fApEn significantly improved the detection accuracy of OSA,and it is an effective detection index to identify the different disease states of healthy people and OSA patients when they are sick or not.

关 键 词：阻塞性睡眠呼吸暂停 短时心率变异性 滑动趋势模糊近似熵 疾病状态分析 

分 类 号：R318[医药卫生—生物医学工程]