On The Analysis of a Compound Neural Network for Detecting Atrio Ventricular Heart Block (AVB) in an ECG Signal

Heart failure is the most common reason of death nowadays, but if the medical help is given directly, the patient-s life may be saved in many cases. Numerous heart diseases can be detected by means of analyzing electrocardiograms (ECG). Artificial Neural Networks (ANN) are computer-based expert systems that have proved to be useful in pattern recognition tasks. ANN can be used in different phases of the decision-making process, from classification to diagnostic procedures. This work concentrates on a review followed by a novel method. The purpose of the review is to assess the evidence of healthcare benefits involving the application of artificial neural networks to the clinical functions of diagnosis, prognosis and survival analysis, in ECG signals. The developed method is based on a compound neural network (CNN), to classify ECGs as normal or carrying an AtrioVentricular heart Block (AVB). This method uses three different feed forward multilayer neural networks. A single output unit encodes the probability of AVB occurrences. A value between 0 and 0.1 is the desired output for a normal ECG; a value between 0.1 and 1 would infer an occurrence of an AVB. The results show that this compound network has a good performance in detecting AVBs, with a sensitivity of 90.7% and a specificity of 86.05%. The accuracy value is 87.9%.




References:
[1] Encyclopédie Médico-Chirurgicale 11-003-F-30, 2004.
[2] P. Baldi and S. Brunak, "Bioinformatics," The Machine Learning
Approach, MIT Press, London, UK, 2000.
[3] D. L. Hudson and M. E. Cohen, "Neural Networks and Artificial
Intelligence," for Biomedical Engineering. Piscataway, NJ: IEEE Press,
2000.
[4] M. N. Liebman, "Biomedical Informatics: the Future for Drug
Development," Drug Discovery Today, Vol. 7, No. 20 (Suppl.), pp. 197-
203, 2002.
[5] De Dombal, F. T., Leaper, D. J., Tanisland, J. R., McCann, A. P., &
Horrocks, J. C. Computer-aided diagnosis of acute abdominal pain.
BMJ, Vol. 2, pp 9-13,1972.
[6] Goovaerts H.G. et al.- A Digital QRS Detector Based on the Principle
of Contour Limiting, IEEE trans. BME. March. 1976.
[7] Rantaharju P.M et al.- The Current State of Computer ECG Analysis: A
Critique Proceeding of Trends in Computer-Processed
Electrocardiograms, 1977.
[8] Okada M.- A Digital Filter for the QRS Complex Detection, IEEE, TBME,
Vol. 26, No 12, Dec. 1979.
[9] Milan Horaced et al.- Feature Enhancement in ECG Potential Maps by
Computation of Infinite Medium Potentials, Proceeding from
Computerized Interpretation of Electrocardiogram, April, 1983.
[10] Baxt, W. G. Application of neural networks to clinical medicine.
Lancet, 346, pp 1135-1138. , 1995
[11] Baxt, W. G. Artificial neural network to identify acute myocardial
infarction-Reply. Lancet, 347, pp 551, 1996.
[12] Baxt, W. G., & Skora, J. Prospective validation of artificial neural
network trained to identify acute myocardial infarction. Lancet, 347, pp
12-15, 1996.
[13] Baxt, W. G., & White, H. Bootstrapping confidence intervals for clinical
input variable effects in a network trained to identify the presence of
acute myocardial infarction. Neural Computation, No 7 Vol. 3, pp 624-
638, 1995.
[14] A. Casaleggio, M. Morando, S. Ridella "Neural Networks for Automatic
Anomalous QRS Complex Detection," Myocardial Infraction Proc.
Computer in Cardiology, pp. 553-6, 1991.
[15] W. Dassen, R. Mulleneers, B. Bleijlevens, K Den Dulk, LM. Rrodriguez,
J. Schlapfer, A. Katsivas, H. wellens, "Determination of the Aetiology of
Wide-QRS Tachycardia Using an Artificial Neural Network," IEEE, pp.
165-68, 1992.
[16] T. Conde, "Automatic Neural Detection of Anomalies in
Electrocardiogram (ECG) Signals," IEEE, pp. 3552-58, 1994.
[17] Y. Ming-Chuan, B. Pariseau, J.M Jenkins, L.A DiCarlo, "Iteracardic
Arrhythmia Classification using Neural Networks and Time Frequency
Analysis," Proc. Computer in Cardiology, pp. 449-52,1994.
[18] M. Hoher, H.A. Kestler, S. Bauer, P. Weismuller, G. Palm, V.
Hombach, "Neural Networks Based Analysis of the Signal Averaged
Electrocardiogram," Proc. Computer in Cardiology, pp. 257-60, 1995.
[19] Ellenius, J., Groth, T., & Lindahl, B. Neural network analysis of
biochemical markers for early assessment of acute myocardial infarction,
Stud. Health Technol. Inform., No 43, pp 382-385, 1997.
[20] B. Heden, L. Edenbrant, W.K. Haisty JR, O. Pahlm, "Neural Networks
for ECG Diagnosis of Inferior Myocardial Infraction," Proc. Computer
in Cardiology, pp. 345-47, 1993.
[21] B. Heden, K. Edenbrant, L. Wesley, W.K. Haisty, O. Pahlm, " Artificial
Neural Networks for Electrocardigraphic Diagnosis of Healed
Myocardial Infraction," The American Journal of Cardiology, Vol. 74,
pp.5-8, 1994.
[22] B. Heden, R. Hans Ohlin, L. Rittner, L. Edenbrant, "Acute Myocardial
Infraction Detected 12-Lead ECG by Artificial Neural, Networks,"
Neural, Networks for ECG Analysis Circulation Vol. 96, No 6, pp.
1798-1802, 1997.
[23] Baxt, W. G. Improving the accuracy of an artificial neural network using
multiple differently trained networks. Neural Computation, No 4 Vol. 5,
, pp 772-780, 1992.
[24] Baxt, W. G. A neural-network trained to identify the presence of
myocardial-infarction bases some decisions on clinical associations that
differ from accepted clinical teaching. Med. Dec. Making, No 14 Vol. 3,
1994, pp 217-222.
[25] Baxt, W. G. A neural-network trained to identify the presence of
myocardial-infarction bases some decisions on clinical associations that
differ from accepted clinical teaching. Med. Dec. Making, No 14, Vol. 3,
pp 217-222, 1994.
[26] Dreiseitl, S., Ohno-Machado, L., & Vinterbo, S. Evaluating variable
selection methods for diagnosis of myocardial infarction. Proc. AMIA
Symp., Vol. 1-2, pp 246-250, 1999.
[27] Neal, R. M. Bayesian learning for neural networks, New York:
Springer-Verlag., 1996.
[28] Selker, H. P., Griffith, J. L., Patil, S., Long, W. J., & D'Agostino, R. B.
A comparison of performance of mathematical predictive methods for
medical diagnosis: identifying acute cardiac ischemia among emergency
department patients. J. Investig. Med., No 43, Vol. 5. pp 468-476, 1995.
[29] Polak, M. J., Zhou, S. H., Rautaharju, P. M., Armstrong, W. W., &
Chaitman, B. R. Using automated analysis of the resting twelve-lead
ECG to identify patients at risk if developing transient myocardial
ischemia-an application of an adaptive logic network. Physiol. Meas.,
No 18 Vol. 4, pp 317-32, 19975.
[30] W. Haiying, F. Azuaje, and B. Norman, "An Integrative and Interactive
Framework for Improving Biomedical Pattern Discovery and
Visualization", IEEE Transaction on Information Technology in
Biomedicine, Vol. 8, No. 1, pp. 16-27, 2004,.
[31] Abdulnasir Hossen, "Biomedical Signals Classification Intelligent
Systems", Proc. CSIT Jordan, 2006.
[32] A. R. Houghton, D. Gray, Ma├«triser L-ECG de la Théorie ├á la Clinique ,
PP. 1-30, Masson,2000
[33] New York Heart Association Subcommittee on Electrocardiography
Criteria, Nomenclature and Criteria for Diagnosis of Diseases of the
Heart and Blood Vessels, 5th ed. 1953.
[34] P. M. Rantahaju et al, The Impact of Computer on Electrocardiogram
Interpretation Systems, NARDP Grants N┬░ 410, 1978.
[35] Schwartz, W. B. Medicine and the computer: the promise and problems
of change. New Engl. J. Med., 283, pp 1257-1264, 1970.