Abstract
Electrocardiography (ECG) is a medical test used to measure the heart’s electrical conduction system. Many cardiac abnormalities can be detected through ECG analysis. Various computerized techniques have been applied to assist physicians in an accurate diagnosis, among them Artificial neural networks (ANNs), Genetic algorithms (GAs) and their combinations (GANNs). A cardiac arrhythmia computerized diagnostics is a good example of such an application. Many different ANN approaches to arrhythmia classification appear in the literature, but we couldn’t find any work related to GANN for this problem. In this paper we are closing this gap by presenting classification system of cardiac arrhythmia using GANN classifier. ANN is the base of the system, while GAs are used to evolve ANN architecture and weights. In addition, another GA is used to perform a feature selection task. The system is trained and tested for online UCI Machine Learning data set for cardiac arrhythmia. The classification performance of the system is evaluated by means of classification accuracy. The proposed classifier gives best classification results of 90.23%, 100%, 94.98%, 99.46%, 97.88% and 86.5% on classifying ischemic changes, old anterior myocardial infarction, old inferior myocardial infarction, sinus tachycardia, sinus bradycardia and right bundle branch block respectively. These results are competitive with the state of the art results in the field; that proves the effectiveness of our application. In addition the tool is enough generic to be used in solving of a wide range of problems. Also we have investigated effectiveness of GA as a training method. Exhaustive experiments demonstrate that classification accuracy of GA-trained classifiers is inversely proportional to the number of classification cases and depends on the content and size of the feature set for the classifier that is built by help of it.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Anuradha, B., Veera Reddy, V.C.: ANN for classification of cardiac arrhythmias. ARPN Journal of Engineering and Applied Sciences 3(3) (2008)
Arifovic, J., Gencay, R.: Using genetic algorithms to select architecture of a feed-forward artificial neural network. Physica A 289, 574–594 (2001)
Begg, R., Kammruzzaman, J., Sarker, R.: Neural networks in healthcare, potential and challenges. IDEA Group (2006)
Bocko, J., Nohajová, V., Harčarik, T.: Application of methods of selection and crossover to identification of parameters of Borden-Partom model. Modeling of mechanical and mechatronic system (2011)
Bronis, K., Kappos, K., Manolis, A.S.: Early repolarization not benign anymore J-wave syndrome. Hospital Chronicles 7(4), 215–228 (2012)
Chawla, N.V., Bowyer, K.W., Hall, L.O., Kegelmeyer, W.P.: SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research 16, 321–357 (2002)
Cox, A.: SADS – sudden arrhythmic death syndrome, Cardiac Risk in the Young. Produced by Cardiac Risk in the Young (2003)
Dokur, Z., Olmez, T.: ECG beat classification by a novel hybrid neural network. Computer Methods and Programs in Biomedicine 66, 167–181 (2001)
Güvenir, A.H., Acar, B., Muderrisoglu, H.: Arrhythmia Data set. Center for Machine Learning and Intelligent Systems. University of California, Irvine
Güvenir, H.A., Acar, B.: Feature selection using a genetic algorithm for the detection of abnormal ECG recordings. In: Proceedings of the World Conference on Systemics, Cybernetics and Informatics (ISAS/SCI 2001), Orlando, Florida, pp. 437–442 (July 2001)
Fiszelew, A., Britos, P., Ochoa, A., Merlino, H., Fernández, E., García-Martínez, R.: Finding optimal network architecture using genetic algorithms. Research in Computing Science 27, 15–24 (2007)
Hall, M.A., Smith, L.A.: Feature Selection for Machine Learning: Comparing a Correlation-based Filter Approach to the Wrapper. American Association for Artificial Intelligence (1998)
Haykin, S.: Neural Networks: a comprehensive foundation. Prentice Hall, Upper Saddle River (1999)
Heaton, J.: Introduction to neural networks for C#, 2nd edn. First Printing. Heaton Research (2008)
Heaton, J.: Programming neural networks with Encog 2 in C#. Heaton Research (2010)
Jadhav, S.M., Nalbalwar, S.L., Ghatol, A.A.: Generalized feed-forward neural network based cardiac arrhythmia classification from ECG signal. In: 2010 6th International Conference on Advanced Information Management and Service (IMS), pp. 351–256 (2010)
Jadhav, S.M., Nalbalwar, S.L., Ghatol, A.A.: Arrhythmia disease classification using artificial neural network model. In: 2010 IEEE International Conference Computational Intelligence and Computing Research (ICCIC), pp. 1–4 (2010)
Kavitha, K.S., Ramakrishnan, K.V., Singh, M.K.: Modeling and design of evolutionary neural network for heart disease detection. International Journal of Computer Science Issues (IJCSI) 7(5) (2010)
Ladha, L., Deepa, T.: Feature Selection Methods and Algorithms. International Journal on Computer Science and Engineering (IJCSE) 3(5) (2011)
Miller, G.F., Todd, P.M., Hegde, S.U.: Designing neural networks using genetic algorithms. In: Schaffer, J.D. (ed.) Proceedings of the Third International Conference on Genetic Algorithms, pp. 379–384. Morgan Kaufmann (1989)
Milley, A.H., Seabolt, J.D., Williams, J.S.: Data Mining and the Case for Sampling. SAS Institute Inc. (1988)
Mitchell, M.: An introduction to genetic algorithms (complex adaptive systems). MIT Press (1996)
Montana, D.J.: Neural network weight selection using genetic algorithms. Intelligent Hybrid Systems (1995)
Nandy, S., Sarkar, P.P., Das, A.: An Improved Gauss-Newtons Method based Back-propagation algorithm for fast convergence. International Journal of Computer Applications (0975 – 8887) 39(8) (2012)
Osowski, S., Siwek, K., Siroic, R.: Neural system for heart beats recognition using genetically integrated ensemble of classifiers. Computers in Biology and Medicine 41, 173–180 (2011)
Raut, R.D., Dudul, S.V.: Arrhythmia classification using MLP neural network and statistical analysis. Emerging Trends in Engineering and Technology (ICETET 2008), 553–558 (2008)
Schaffer, J.D., Whitley, D., Eshelman, L.J.: Combinations of genetic algorithms and neural networks: a survey of the state of the art. International Workshop on COGANN 92, 1–37 (1992)
Sekkal, M., Chikh, M.A., Settouti, N.: Evolving neural networks using a genetic algorithm for heartbeat classification. Journal of Medical Engineering and Technology 35(5), 215–223 (2011)
Somasundaram, R.S., Nedunchezhian, R.: Evaluation of three simple imputation methods for enhancing preprocessing of data with missing values. International Journal of Computer Applications 21(10), 14 (2011)
Staroszczyk, T., Osowski, S., Markiewicz, T.: Comparative Analysis of Feature Selection Methods for Blood Cell Recognition in Leukemia. In: Perner, P. (ed.) MLDM 2012. LNCS, vol. 7376, pp. 467–481. Springer, Heidelberg (2012)
Taskiran, H.D.: A genetic algorithm approach to feature subset selection for pattern classification using neural networks. Intelligent Systems and their Applications (2005)
University of Pittsburg medical center, Arrhythmia (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Maliarsky, E., Avigal, M., Herman, M. (2014). A Neuro-Genetic System for Cardiac Arrhythmia Classification. In: Perner, P. (eds) Machine Learning and Data Mining in Pattern Recognition. MLDM 2014. Lecture Notes in Computer Science(), vol 8556. Springer, Cham. https://doi.org/10.1007/978-3-319-08979-9_26
Download citation
DOI: https://doi.org/10.1007/978-3-319-08979-9_26
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08978-2
Online ISBN: 978-3-319-08979-9
eBook Packages: Computer ScienceComputer Science (R0)