Design of a Unified Framework for Analyzing Long-Duration Ambulatory ECG: Application for Extracting QRS Geometrical Features

Purpose Since ambulatory electrocardiogram (ECG) signal is always accompanied by strong noise, high amplitude baseline wandering, impulsive artifacts, arrhythmic outliers and some discontinuities, these effects reduce the accuracy of a computerized cardiac-originated events detection-delineation algorithm. The aim of this study is to describe a multi-aspect robust structure of a solution designed for detection-delineation of major events of the long-duration holter ECG signal. Methods In this work, after application of appropriately adopted preprocessing steps, a uniform-length sliding window was moved sample to sample on the preprocessed signal. In each slid, six geometrical features of the excerpted segment were calculated aimed for generating the newly defined geometric index (GI) metric. Then, the alpha-level Neyman-Pearson classifier was designed and implemented to detect and delineate QRS events. Results The presented method was applied to diverse number of databases and as a result, the average values of sensitivity Se = 99.96% and positive predictivity P+ = 99.96% were obtained for the detection of QRS complexes, with the average maximum delineation error 5.7, 3.8 and 6.1 msec for P-wave, QRS complex and T-wave, respectively. Also, the proposed method was applied to DAY general hospital high-resolution holter data and the average values of Se=99.98% and P+=99.97% were obtained for QRS detection. Conclusions It is observed that the proposed method successfully detects and delineates the ECG events showing marginal improvement of the ECG events detection-delineation recent studies.