Implementation of Algorithms for Identification of QRS-Complexes in Real-Time ECG Systems

Abstract

Many modern methods of computer electrocardiography involve identification of QRS-complexes by determining the position of their maximums (R-wave). The obtained sequence of R-waves is used for segmentation of the cardiocycle, measurement of amplitude and duration of its elements, determination of the isoline by the ECG method of polynomial approximation, and analysis of the cardiac rhythm variability. Realtime ECG systems provide continuous processing of ECG during recording of the signal. Such systems should rapidly detect all changes in the characteristics of the ECG signal (heart rate, fibrillation, etc.). This imposes stringent requirements on the algorithms for processing of ECG information. The following requirements for the algorithms are especially important: 1) high productivity matching the input data flow rate; 2) efficacious operation during a specified time interval; 3) capability for efficacious operation in the case of limited availability of input data. Fulfillment of the first and second requirements depends on the performance of ECG hardware and software, whereas the third requirement is imposed on the processing algorithm itself.