Title :
An open-source method for simulating atrial fibrillation using ECGSYN
Author :
Healey, J. ; Clifford, G.D. ; Kontothanassis, L. ; McSharry, P.E.
Author_Institution :
Hewlett-Packard Lab., Cambridge, MA, USA
Abstract :
A method for simulating atrial fibrillation has been developed using the open source simulator ECGSYN (available at physionet.org). In this method episodes of atrial fibrillation are simulated by combining data streams generated by two ECGSYN engines driven from a correlated stochastic process. The underlying process represents atrial activity during fibrillation. This process generates a series of atrial inter-beat intervals with the same statistical properties as those found in the MIT-BIH atrial fibrillation database (afdb). The atrial beats from this process are then selected to be either conducting beats that generate subsequent ventricular activity or nonconducting beats that generate only atrial activity. The intervals for conducting beats are processed by an ECGSYN engine with appropriate parameters to create PQRS and T waves while the nonconducting intervals are processed by a second ECGSYN engine with parameters to generate only P waves. The data streams from the two engines are superimposed to create an artificial atrial fibrillation waveform. This waveform generator has been made into an operator and has been incorporated into a stream based ECG simulator. The simulator uses a timing operator to switch from generating normal ECG morphologies to atrial fibrillation.
Keywords :
diseases; electrocardiography; medical signal processing; stochastic processes; waveform generators; ECG morphology; ECGSYN engine; MIT-BIH atrial fibrillation database; artificial atrial fibrillation waveform; atrial activity; atrial fibrillation; atrial inter-beat interval; correlated stochastic process; open source simulator; statistical properties; stream based ECG simulator; timing operator; ventricular activity; waveform generator; Atrial fibrillation; Databases; Electrocardiography; Engines; Morphology; Open source software; Signal generators; Stochastic processes; Switches; Timing;
Conference_Titel :
Computers in Cardiology, 2004
Print_ISBN :
0-7803-8927-1
DOI :
10.1109/CIC.2004.1442964