机电工程技术
機電工程技術
궤전공정기술
MACHANICAL & ELECTRICAL ENGINEERING TECHNOLOGY
2013年
8期
106-110
,共5页
心电预处理%心电特征检测%自适应阈值差分法%局域变换法
心電預處理%心電特徵檢測%自適應閾值差分法%跼域變換法
심전예처리%심전특정검측%자괄응역치차분법%국역변환법
pre-processing of ECG%ECG feature detection%adaptive slope threshold method%local transformation method
提出一种针对心电信号特征检测的快速算法。首先针对干扰信号的频率特性分别采用了中值滤波、梳状滤波器、Butter-worth滤波器去除噪声干扰。然后根据心电波形特性采用一种自适应阈值差分算法来进行QRS波检测。接着采用改进后的局域变换法检测P、T波。利用MIT-BIH的QT数据库对所提出的算法进行评估，QRS群波检测率达99.65%以上，对P、T波的定位误差在50 ms以内。试验结果表明，该算法能提高各波的检测精度，适于实际应用。
提齣一種針對心電信號特徵檢測的快速算法。首先針對榦擾信號的頻率特性分彆採用瞭中值濾波、梳狀濾波器、Butter-worth濾波器去除譟聲榦擾。然後根據心電波形特性採用一種自適應閾值差分算法來進行QRS波檢測。接著採用改進後的跼域變換法檢測P、T波。利用MIT-BIH的QT數據庫對所提齣的算法進行評估，QRS群波檢測率達99.65%以上，對P、T波的定位誤差在50 ms以內。試驗結果錶明，該算法能提高各波的檢測精度，適于實際應用。
제출일충침대심전신호특정검측적쾌속산법。수선침대간우신호적빈솔특성분별채용료중치려파、소상려파기、Butter-worth려파기거제조성간우。연후근거심전파형특성채용일충자괄응역치차분산법래진행QRS파검측。접착채용개진후적국역변환법검측P、T파。이용MIT-BIH적QT수거고대소제출적산법진행평고，QRS군파검측솔체99.65%이상，대P、T파적정위오차재50 ms이내。시험결과표명，해산법능제고각파적검측정도，괄우실제응용。
This paper describes a fast algorithm for feature detection of ECG. Firstly, the frequency characteristics of the interference signal with a median filter,comb filter,Butterworth filter is to remove interference. Then according to the ECG waveform characteristics adopt an adaptive threshold difference algorithm to detect QRS wave. Then use improved local transformation method to detect P,T wave. The data from the MIT-BIH QT database is validated to verify the improved algorithm, detection rate of the QRS complexes excelled to 99.65. The errors of the P-wave position and T-wave position are reduced under 50 milliseconds or less. Experimental results show that the improved algorithm improves the detection accuracy of the waves,accurate and suitable for practical application.
