An Efficient Instantaneous ECG Delineation Algorithm

Yuan Wen Hau1, Thion Ming Chieng2, Zaid Omar1, Chiao Wen Lim1
1Dr., 2Mr.


Abstract

An efficient electrocardiogram (ECG) delineation algorithm based on discrete wavelet transform and moving window average techniques is proposed to instantaneously delineate the ECG characteristic points, such as peak, onset and offset points of QRS, P and T waves. Through delineation, most of the clinical significant features can be derived from ECG morphologies. Thus, it is essential to delineate the ECG characteristic waves accurately and precisely as it ensure the performance of the ECG analysis and diagnosis. The proposed delineation algorithm is based on discrete wavelet transform (DWT) and moving window average (MWA) techniques, where the DWT is implemented to de-noise ECG artifacts and separated the de-noised signal into two isolated signal, comprises of different ECG characteristic wave based on their frequency range through DWT decomposition and reconstruction. Two MWA with different window size are implemented to localize the interest window of QRS, P and T waves and consequently delineate its peak. The window size of MWA are adapted to the duration of targeted ECG wave. The onset and offset points are identified as turning or inflection point with first and second differentiation methods within predefined window. The proposed delineation algorithm is evaluated and validated with the modified Lead II data of QT database in term of its accuracy, sensitivity and positive predictive value. With the only 13 sets QT database records with modified Lead II data, the proposed algorithm achieved significant delineation performance with the accuracy above 90.82% for all ECG characteristic points, with R peak achieved best accuracy of 99.8046. As opposite, delineation of T offset point only achieved the accuracy of 86.32%. The time required to delineate 15 minute-long ECG record is only 74.702 second. As conclusion, the proposed ECG delineation algorithm based on DWT and moving window average techniques have been proven simple, accurate and efficient.