TY - JOUR
T1 - Fractal QRS-complexes pattern recognition for imperative cardiac arrhythmias
AU - Lin, Chia Hung
AU - Du, Yi Chun
N1 - Funding Information:
This work is supported in part by the National Science Council of Taiwan under contract no. NSC97-2221-E-244-001 (August 1 2008–July 31 2009). The authors also thank associate editor of Digital Signal Processing and reviewers for reviewing the manuscript and providing the suggestion.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2010/7
Y1 - 2010/7
N2 - This paper proposes using fractal QRS-complexes pattern recognition for imperative cardiac arrhythmias. A typical electrocardiogram (ECG) signal is comprised of P-wave, QRS-complex, and T-wave. Fractal dimension transformation (FDT) is employed to adjoin the QRS-complex from time-domain ECG signals, including the fractal features of supraventricular ectopic beat, bundle branch ectopic beat, and ventricular ectopic beat. FDT with fractal dimension (FD) is addressed for constructing various symptomatic features, and can produce family functions and enhance features, making the difference between healthy and unhealthy subjects more significant. The probabilistic neural network (PNN) is proposed for recognizing the states of cardiac physiologic function. The proposed method is tested using the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) arrhythmia database. Compared with other methods, the numerical experiments demonstrate greater efficiency and higher accuracy in recognizing ECG signals.
AB - This paper proposes using fractal QRS-complexes pattern recognition for imperative cardiac arrhythmias. A typical electrocardiogram (ECG) signal is comprised of P-wave, QRS-complex, and T-wave. Fractal dimension transformation (FDT) is employed to adjoin the QRS-complex from time-domain ECG signals, including the fractal features of supraventricular ectopic beat, bundle branch ectopic beat, and ventricular ectopic beat. FDT with fractal dimension (FD) is addressed for constructing various symptomatic features, and can produce family functions and enhance features, making the difference between healthy and unhealthy subjects more significant. The probabilistic neural network (PNN) is proposed for recognizing the states of cardiac physiologic function. The proposed method is tested using the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) arrhythmia database. Compared with other methods, the numerical experiments demonstrate greater efficiency and higher accuracy in recognizing ECG signals.
UR - http://www.scopus.com/inward/record.url?scp=77955323408&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955323408&partnerID=8YFLogxK
U2 - 10.1016/j.dsp.2009.12.005
DO - 10.1016/j.dsp.2009.12.005
M3 - Article
AN - SCOPUS:77955323408
SN - 1051-2004
VL - 20
SP - 1274
EP - 1285
JO - Digital Signal Processing: A Review Journal
JF - Digital Signal Processing: A Review Journal
IS - 4
ER -