Srinivasan Jayaraman, Adithya V
Annu Int Conf IEEE Eng Med Biol Soc. 2015;2015:5924-7. doi: 10.1109/EMBC.2015.7319740.
Electroencephalogram (EEG) signal artifacts are caused by various factors, such as, Electro-oculogram (EOG), Electromyogram (EMG), Electrocardiogram (ECG), movement artifact and line interference. The relatively high electrical energy cardiac activity causes EEG artifacts. In EEG signal processing the general approach is to remove the ECG signal. In this paper, we introduce an automated method to extract the ECG signal from EEG using wavelet and Teager-Kaiser energy operator for R-peak enhancement and detection. From the detected R-peaks the heart rate (HR) is calculated for clinical diagnosis. To check the efficiency of our method, we compare the HR calculated from ECG signal recorded in synchronous with EEG. The proposed method yields a mean error of 1.4% for the heart rate and 1.7% for mean R-R interval. The result illustrates that, proposed method can be used for ECG extraction from single channel EEG and used in clinical diagnosis like estimation for stress analysis, fatigue, and sleep stages classification studies as a multi-model system. In addition, this method eliminates the dependence of additional synchronous ECG in extraction of ECG from EEG signal process.
脑电图（EEG）信号伪迹由多种因素引起，例如眼电图（EOG）、肌电图（EMG）、心电图（ECG）、运动伪迹和线路干扰。相对较高的心脏电活动能量会导致脑电图伪迹。在脑电图信号处理中，一般方法是去除心电图信号。在本文中，我们介绍一种自动方法，利用小波和Teager-Kaiser能量算子从脑电图中提取心电图信号，用于R波峰增强和检测。根据检测到的R波峰计算心率（HR）以用于临床诊断。为检验我们方法的有效性，我们将从与脑电图同步记录的心电图信号计算得到的心率进行比较。所提出的方法心率平均误差为1.4%，平均R-R间期平均误差为1.7%。结果表明，所提出的方法可用于从单通道脑电图中提取心电图，并作为多模型系统用于临床诊断，如压力分析、疲劳和睡眠阶段分类研究的评估。此外，该方法在从脑电图信号过程中提取心电图时消除了对额外同步心电图的依赖。
