基于心电信号的睡眠呼吸暂停综合征检测算法

An Algorithm Based on ECG Signal for Sleep Apnea Syndrome Detection

睡眠呼吸暂停综合征(SAS)的诊断对预防高血压、冠心病、心律失常、脑卒中等疾病具有重要的临床意义。本文基于单通道的心电(ECG)信号提出了一种新的SAS检测算法,该算法首先对ECG信号进行QRS波检测和预处理,得到RR间期和心源性呼吸(EDR)信号序列,并从中提取各项时频特征参数(n=40),再对这些参数归一化处理后,使用支持向量机(SVM)进行分类训练,最终获得可评判SAS的指标。经MIT-BIH权威数据库(Apnea-ECG database)实验表明,本文算法在训练集和测试集上的准确率分别为95%和88%。

The diagnosis of sleep apnea syndrome （SAS） has a significant importance in clinic for preventing diseases of hypertention, coronary heart disease, arrhythmia and cerebrovascular disorder, etc. This study presents a novel method for SAS detection based on single-channel electrocardiogram （ECG） signal. The method preprocessed ECG and detected QRS waves to get RR signal and ECG-derived respiratory （EDR） signal. Then 40 time- and spectral-do- main features were extracted to normalize the signals. After that support vector machine （SVM） was used to classify the signals as ＂apnea＂ or ＂normal＂. Finally, the performance of the method was evaluated by the MIT-BIH Apnea- ECG database, and an accuracy of 95% in train sets and an accuracy of 88% in test sets were achieved.