Prediction of cholesterol level in patients with myocardial infarction based on medical data mining methods

Authors

  • Cemil Colak Dept. of Biostatistics and Medical Informatics, Inonu University, Faculty of Medicine, Malatya, Turkey
  • Mehmet C. Colak Dept. of Cardiovascular Surgery, Inonu University, Faculty of Medicine, Malatya, Turkey
  • Necip Ermis Dept. of Cardiology, Inonu University, Faculty of Medicine, Malatya, Turkey
  • Nevzat Erdil Dept. of Cardiovascular Surgery, Inonu University, Faculty of Medicine, Malatya, Turkey
  • Ramazan Ozdemir Dept. of Cardiology, Inonu University, Faculty of Medicine, Malatya, Turkey

Keywords:

Artificial neural networks (ANNs), cholesterol level, medical data mining, myocardial infarction (MI), support vector machine (SVM).

Abstract

Myocardial infarction (MI) is a significant reason for death and disability over the world and might be the first signof coronary artery disease. The current study was carried out to predict the cholesterol level in patients with MI usingdata mining methods, artificial neural networks (ANNs) and support vector machine (SVM) models. The data of 596patients, who had been diagnosed with segment elevation MI were analysed in the present study. The retrospectivedataset including gender, age, weight, height, pulse, glucose, creatinine, triglyceride, high-density lipoprotein, andlow-density lipoprotein was used for predicting the cholesterol level. Correlation based feature selection was applied.Multilayer perceptron (MLP) ANNs and SVM with radial basis function kernel were used for the prediction basedon the selected predictors. The performance of the ANNs and SVM models was evaluated on the basis of correlationcoefficient and mean absolute error. The estimated correlation coefficients observed and predicted values were 0.94 forANNs and 0.88 for SVM in training dataset (n=376), and 0.95 for ANNs and 0.90 for SVM in testing dataset (n=160),respectively. ANNs and SVM models yielded mean absolute error of 7.37 and 14.18 in training dataset, and 7.87 and14.71 in testing dataset, consecutively. The results of the performance evaluation showed that MLP ANNs performedbetter for the prediction of cholesterol level in patients with MI in comparison to SVM. The proposed MLP ANNs modelmight be employed for predicting the level of cholesterol for MI patients in clinical decision support process.

References

Arslan, A.K., Colak, C. & Sarihan, M.E. (2016). Different medical

data mining approaches based prediction of ischemic stroke. Computer

Methods and Programs in Biomedicine, 130:87-92.

Colak, C., Karaman, E. & Turtay, M.G. (2015). Application of

knowledge discovery process on the prediction of stroke. Computer

Methods and Programs in Biomedicine, 119(3):181-185.

Colak, M.C., Colak, C., Kocatürk, H., Sağiroğlu, S. & Barutçu,

I. (2008). Predicting coronary artery disease using different artificial

neural network models. Anadolu kardiyoloji dergisi: AKD= the

Anatolian Journal of Cardiology, 8(4):249-254.

Fayyad, U., Piatetsky-Shapiro, G. & Smyth, P. (1996). From data

mining to knowledge discovery in databases. AI Magazine, 17(3):37.

Hongfei, W., Yunyan, Z., Fei, Y. & Hui, L. (2013). Evaluation of an

artificial neural network to ascertain why there is a high incidence of

hepatitis B in the Chinese population after vaccination. Computers in

Biology and Medicine, 43(9):1167-1170.

Investigators, L.R.C. (1992). The lipid research clinics coronary

primary prevention trial: results of 6 years of post-trial follow-up.

Archives of Internal Medicine, 152(7):1399.

Moses, D. (2015). A survey of data mining algorithms used in

cardiovascular disease diagnosis from multi-lead ECG data. Kuwait

Journal of Science, 42(2):206-235.

Shin, J., Park, H., Cho, S., Nam, H. & Lee, K.J. (2014). A correction

method using a support vector machine to minimize hematocrit

interference in blood glucose measurements. Computers in Biology and

Medicine, 52(0):111-118.

Silwattananusarn, T. & Tuamsuk, K. (2012). Data mining and its

applications for knowledge management: A literature review from

to 2012. International Journal of Data Mining & Knowledge

Management Process, 2(5):13-24.

Stamler, J., Wentworth, D. & Neaton, J.D. (1986). Is relationship

between serum cholesterol and risk of premature death from coronary

heart disease continuous and graded?: Findings in 356 222 primary

screenees of the multiple risk factor intervention trial (MRFIT). Jama,

(20):2823-2828.

Süt, N. & Çelik, Y. (2012). Prediction of mortality in stroke patients

using multilayer perceptron neural networks. Turkish Journal of

Medical Sciences, 42(5):886-893.

Thygesen, K., Alpert, J.S., Jaffe, A.S., White, H.D., Simoons, M.L.,

et al.(2012). Third universal definition of myocardial infarction. Journal

of the American College of Cardiology, 60(16):1581-1598.

Thygesen, K., Alpert, J.S. & White, H.D. (2007a). Universal definition

of myocardial infarction. European Heart Journal, 28(20):2525-2538.

Thygesen, K., Alpert, J.S., White, H.D., Jaffe, A.S., Apple, F.S., et

al. (2007b). Universal definition of myocardial infarction. Circulation,

(22):2634-2653.

Wang, S.J., Ohno-Machado, L., Fraser, H.S.F. & Kennedy, R.L.

(2001). Using patient-reportable clinical history factors to predict

myocardial infarction. Computers in Biology and Medicine, 31(1):1-13.

Wilson, P.W., D’Agostino, R.B., Levy, D., Belanger, A.M.,

Silbershatz, H., et al. (1998). Prediction of coronary heart disease

using risk factor categories. Circulation, 97(18):1837-1847.

Yu, W., Liu, T., Valdez, R., Gwinn, M. & Khoury, M.J. (2010).

Application of support vector machine modeling for prediction of

common diseases: the case of diabetes and pre-diabetes. BMC Medical

Informatics and Decision Making, 10(1):16.

Yusuf, S., Hawken, S., Ôunpuu, S., Dans, T., Avezum, A., et al.

(2004). Effect of potentially modifiable risk factors associated with

myocardial infarction in 52 countries (the INTERHEART study): casecontrol

study. The Lancet, 364(9438):937-952.

Zhou, S., Li, G.B., Huang, L.Y., Xie, H.Z., Zhao, Y.L., et al. (2014).

A prediction model of drug-induced ototoxicity developed by an

optimal support vector machine (SVM) method. Computers in Biology

and Medicine, 51:122-127.

Downloads

Published

08-08-2016