Diagnosis of Heart Disease Based on Meta Heuristic Algorithms and Clustering Methods

Document Type: Research Paper

Authors

Islamic Azad University, Ferdows Branch, ferdows, Iran.

Abstract

Data analysis in cardiovascular diseases is difficult due to large massive of information. All of features are not impressive in the final results. So it is very important to identify more effective features. In this study, the method of feature selection with binary cuckoo optimization algorithm is implemented to reduce property. According to the results, the most appropriate classification for support vector machine is featured diagnoses heart disease. The main purpose of this article is feature reduction and providing a more precise diagnosis of the disease. The proposed method is evaluated using three measures: accuracy, sensitivity and specificity. For comparison, a data set of Machine Learning Repository database including information about 303 people with 14 features was used. In addition to the high accuracy of current methods, are expensive and time-consuming. The results indicate that the proposed method is superior on other algorithms in terms of Performance, accuracy and run time.

Keywords


[1] MAYO CLINIC. 2014. Diseases and Conditions Heart disease. http://www.mayoclinic.org/diseasesconditions/heart disease/ basics/ definition/ con20034056.

[2] World Health Organization. 2014. Reviewed June 2016. WHO cardiovascular disease. http://www.who.int/mediacentre/factsheets/fs317/en/.

[3] N. Cong Long, P. Meesad, and H. Unger, “A highly accurate firefly based algorithm for heart disease prediction,” Expert Systems with Applications, vol. 42, pp. 8221-8231, 2015.

[4] Z. Mahmoodabai and S. Shaerbaf Tabrizi, “A new ICA-based algorithm for diagnosis of coronary artery disease,” Intelligent Computing, Communication and Devices, vol. 2, pp. 415-427, 2014.

[5] A. Dewan and M. Sharma, “Prediction of heart disease using a hybrid technique in data mining classification,” Computing for Sustainable Global Development (INDIACom), pp. 704-706, 2015.

[6] F. Ahmad, N. Ashidi Mat Isa, Z. Hussain, and M. Khusairi Osman, “Intelligent medical disease diagnosis using improved hybrid genetic algorithm-multilayer perceptron network,” Journal of Medical Systems, vol. 37, pp. 9934, 2013.

[7] https://archive.ics.uci.edu/ml/datasets/Statlog+(Heart).

[8] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning” vol. 20, pp. 273-297, 1995.

[9] S. Mahmoudi, R. Rajabioun, and S. Lotfi, “Binary cuckoo optimization algorithm,” National Conference on New Approaches in Computer Engineering and Information Retrieval, Iran, Roudsar, 2013.

[10] G. Chandrashekar and F. Sahin, “A survey on feature selection methods,” Computers & Electrical Engineering, vol. 40, no. 1, pp. 16-28, 2014.

[11] A. Fadzil, M. Nor Ashidi, H. Zakaria, O. Muhammad Khusairi, and S. Siti Noraini, “A GA-based feature selection and parameter optimization of an ANN in diagnosing breast cancer,” Pattern Analysis and Applications, vol. 14, pp. 861-870, 2014.