A NOVEL APPROACH TO HEART ATTACK PREDICTION IMPROVEMENT VIA EXTREME LEARNING MACHINES CLASSIFIER INTEGRATED WITH DATA RESAMPLING STRATEGY

Abstract

The heart attack is a disorder that is frequently seen in low-income countries and causes the death of many people. Cardiologists benefit from electrocardiography (ECG) tests to determine this condition. Supervised classification algorithms are frequently used and provide very successful results in computer-aided diagnostic systems. In this study, a new approach to predict a heart attack is proposed for classification via extreme learning machines (ELM) integrated with the resampling strategy. This study aims to reveal a new diagnostic system that will increase the success of current studies. The study has three basic steps. In order to determine the features that will ensure the system’s optimized operation, firstly, the ReliefF feature selection method was applied to the data set, and then, the system was modeled by different classifiers via resampling. Besides, the as-proposed approach was applied to the breast cancer data to test the accuracy of the current system. The as-obtained results from both Statlog (heart disease) and the breast cancer data were seemed to be more successful than the studies in the literature. Thus, the as-proposed system reveals a successful and effective approach that can be applied in clinical data sets.

Keywords

• Data resample
• heart attack prediction
• feature selection methods
• statlog data set

VERİ YENİDEN ÖRNEKLEME STRATEJİSİ İLE BÜTÜNLEŞTİRİLMİŞ AŞIRI ÖĞRENME MAKİNELERİ SINIFLAYICILARI İLE KALP KRİZİ TAHMİNLERİNİN İYİLEŞTİRİLMESİ İÇİN YENİ BİR YAKLAŞIM

Öz

Kalp krizi düşük gelirli ülkelerde sık görülen ve birçok insanın ölümüne neden olan bir hastalıktır. Kardiyologlar bu durumu belirlemek için elektrokardiyografi (EKG) testlerinden yararlanırlar. Denetimli sınıflandırma algoritmaları, bilgisayar destekli tanılama sistemlerinde sıklıkla kullanılır ve çok başarılı sonuçlar verir. Bu çalışmada, kalp krizini öngörmede yeniden örnekleme stratejisiyle bütünleşmiş aşırı öğrenme makineleri (ELM) ile yapılan sınıflandırma için yeni bir yaklaşım önerilmiştir. Bu çalışmanın amacı, güncel çalışmaların başarısını artıracak yeni bir tanı sistemi ortaya koymaktır. Çalışmanın üç temel adımı vardır. İlk aşamada, ReliefF özellik seçim yöntemi veri setine uygulanır ve sistemin en iyi şekilde çalışmasını sağlayacak özellikler belirlenir. Daha sonra sistem yeniden örnekleme ile farklı sınıflandırıcılarla modellenmiştir. Ek olarak, önerilen yaklaşım meme kanseri verilerine uygulanmış ve mevcut sistemin doğruluğu test edilmiştir. Hem Statlog (kalp krizi) hem de meme kanseri verilerinin sonuçları literatürdeki çalışmalardan daha başarılı sonuçlar vermiştir. Böylece, önerilen sistem, klinik veri setlerinde uygulanabilecek başarılı ve etkili bir yaklaşım ortaya koymaktadır.

Anahtar Kelimeler

• Veri yeniden örnekleme
• kalp krizi tahmini
• özellik seçim yöntemleri
• Statlog Veri seti

References

1. Buscema, M., Breda, M., & Lodwick, W., 2013, Training With Input Selection and Testing (TWIST) algorithm: a significant advance in pattern recognition performance of machine learning, Journal of Intelligent Learning Systems and Applications, 5(1), 29.
2. Chen, J. Y., & Lonardi, S. 2009, Biological data mining: CRC Press.
3. Das, R., Turkoglu, I., & Sengur, A., 2009, Effective diagnosis of heart disease through neural networks ensembles, Expert Systems with Applications, 36(4), 7675-7680.
4. Das, S., 2001, Filters, wrappers and a boosting-based hybrid for feature selection, Paper presented at the Icml.
5. Dewan, A., & Sharma, M., 2015, Prediction of heart disease using a hybrid technique in data mining classification, Paper presented at the Computing for Sustainable Global Development (INDIACom), 2015 2nd International Conference on.
6. Duch, W., Adamczak, R., & Grabczewski, K., 2001, A new methodology of extraction, optimization and application of crisp and fuzzy logical rules, IEEE Transactions on Neural Networks, 12(2), 277-306.
7. Freeman, C., Kulić, D., & Basir, O., 2015, An evaluation of classifier-specific filter measure performance for feature selection, Pattern Recognition, 48(5), 1812-1826.
8. Guyon, I., & Elisseeff, A., 2003, An introduction to variable and feature selection, Journal of machine learning research, 3(Mar), 1157-1182.

9. Helmy, T., & Rasheed, Z., 2009, Multi-category bioinformatics dataset classification using extreme learning machine, Paper presented at the Evolutionary Computation, 2009. CEC'09. IEEE Congress on.
10. Huang, G.-B., Zhu, Q.-Y., & Siew, C.-K., 2004, Extreme learning machine: a new learning scheme of feedforward neural networks, Neural Networks, 2004., 2, 985-990.
11. Huang, G.-B., Zhu, Q.-Y., & Siew, C.-K., 2006, Extreme learning machine: theory and applications, Neurocomputing, 70(1-3), 489-501.
12. Huang, G. B., Zhou, H., Ding, X., & Zhang, R., 2012, Extreme Learning Machine for Regression and Multiclass Classification, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 42(2), 513-529. doi:10.1109/TSMCB.2011.2168604.
13. Kahramanli, H., & Allahverdi, N., 2008, Design of a hybrid system for the diabetes and heart diseases, Expert Systems with Applications, 35(1), 82-89. doi:https://doi.org/10.1016/j.eswa.2007.06.004.
14. Karabatak, M., 2015, A new classifier for breast cancer detection based on Naïve Bayesian, Measurement, 72, 32-36.
15. Karabatak, M., & Ince, M. C., 2009, An expert system for detection of breast cancer based on association rules and neural network, Expert Systems with Applications, 36(2), 3465-3469.
16. Karabulut, E. M., & İbrikçi, T., 2012, Effective diagnosis of coronary artery disease using the rotation forest ensemble method, Journal of medical systems, 36(5), 3011-3018.
17. Liu, X., Wang, X., Su, Q., Zhang, M., Zhu, Y., Wang, Q., & Wang, Q., 2017, A hybrid classification system for heart disease diagnosis based on the rfrs method, Computational and mathematical methods in medicine, 2017.
18. Liu, X., & Zeng, Z., 2015, A new automatic mass detection method for breast cancer with false positive reduction, Neurocomputing, 152, 388-402.
19. Long, N. C., Meesad, P., & Unger, H., 2015, A highly accurate firefly based algorithm for heart disease prediction, Expert Systems with Applications, 42(21), 8221-8231.
20. Mitra, S., & Pathak, P., 1984, The nature of simple random sampling, The Annals of Statistics, 1536-1542.
21. Nahar, J., Imam, T., Tickle, K. S., & Chen, Y.-P. P., 2013, Computational intelligence for heart disease diagnosis: A medical knowledge driven approach, Expert Systems with Applications, 40(1), 96-104.
22. Organization, W. H. 2017, Cardiovascular diseases (CVDs), Retrieved from https://www.who.int/cardiovascular_diseases/en/
23. Özşen, S., & Güneş, S., 2008, Effect of feature-type in selecting distance measure for an artificial immune system as a pattern recognizer, Digital Signal Processing, 18(4), 635-645. doi:https://doi.org/10.1016/j.dsp.2007.08.004
24. Polat, K., & Güneş, S., 2009, A new feature selection method on classification of medical datasets: Kernel F-score feature selection, Expert Systems with Applications, 36(7), 10367-10373. doi:https://doi.org/10.1016/j.eswa.2009.01.041
25. Robnik-Šikonja, M., & Kononenko, I., 2003, Theoretical and empirical analysis of ReliefF and RReliefF, Machine learning, 53(1-2), 23-69.
26. Saygılı, A., 2018, Classification and Diagnostic Prediction of Breast Cancers via Different Classifiers, International Scientific and Vocational Studies Journal, 2(2), 48-56.
27. Spekowius, G., & Wendler, T. 2006, Advances in healthcare technology: shaping the future of medical care (Vol. 6): Springer Science & Business Media.
28. Subbulakshmi, C., Deepa, S., & Malathi, N., 2012, Extreme learning machine for two category data classification, Paper presented at the Advanced Communication Control and Computing Technologies (ICACCCT), 2012 IEEE International Conference on.
29. Şahan, S., Polat, K., Kodaz, H., & Güneş, S., 2005, The medical applications of attribute weighted artificial immune system (AWAIS): diagnosis of heart and diabetes diseases, Paper presented at the International Conference on Artificial Immune Systems.
30. Takci, H., 2018, Improvement of heart attack prediction by the feature selection methods, Turkish Journal of Electrical Engineering & Computer Sciences, 26(1), 1-10.
31. UCI Machine Learning Repository, Statlog (Heart) Data Set Retrieved from http://archive.ics.uci.edu/ml/datasets/Statlog+%28Heart%29 Wang, P., Hu, X., Li, Y., Liu, Q., & Zhu, X., 2016, Automatic cell nuclei segmentation and classification of breast cancer histopathology images, Signal Processing, 122, 1-13.
32. Wang, S.-j., Mathew, A., Chen, Y., Xi, L.-f., Ma, L., & Lee, J., 2009, Empirical analysis of support vector machine ensemble classifiers, Expert Systems with Applications, 36(3, Part 2), 6466-6476. doi:https://doi.org/10.1016/j.eswa.2008.07.041
33. Yilmaz, N., Inan, O., & Uzer, M. S., 2014, A new data preparation method based on clustering algorithms for diagnosis systems of heart and diabetes diseases, Journal of medical systems, 38(5), 48.