Comparative assessments of the efficiency and noise immunity of the algorithms detecting the shock cardiac rhythms (SCR) are described in the paper. It is suggested to use the rejection filter of ventricular fibrillation as the principle algorithm for the SCR detection in an automated external defibrillator. The use of the method combining the after-threshold interval and the method of spectral analysis is demonstrated to be preferable for a portable external defibrillator.
The paper reviews the existing methods for identifying QRS complexes. An attempt is made to make a well-defined classification of available algorithms in an approach to QRS detection. Emphasis is laid on the consideration of specific requirements for such algorithms by computer-aided real-time ECG systems. A task of choice of the most suitable method is formulated. An algorithm based on the well-known principles of frequency-time detection is proposed as an alternative solution of this task. The algorithm makes it possible to single out QRS complexes from real-time ECG and to effectively make a digital signal processing by available optimized libraries. The method initially used in the Matlab package has been integrated within the laboratory computerized ECG system.
Despite the fact that the duration of cardiac cycle complexes is a main diagnostic index in interpretation of ECG, a unified common approach to constructing measuring algorithms is lacking now. The paper reviews the methods of detection and measurement of the onset and termination of cardiographic complexes. Their link with the general theory of assessment by using the Fischer mathematical statistic methods, the Winner filtration methods, the approximation methods by spline functions and Kalman's filters is considered. An attempt was made to describe the pattern of cardiac cycle pulses on the basis of physiological processes of cardiac electric activity.
Lebedev VV, Kalantar VA, Arakcheev AG, and Korado IV
Subjects
Humans, Algorithms, Electrocardiography, Signal Processing, Computer-Assisted
Abstract
The paper defines requirements for reference signals to check measuring the algorithms of automatic electrocardiographic systems. It proposes a reference signal which meets these requirements and gives a block diagram of a test bed.
Published
1997
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.