Heart attack mortality prediction: an application of machine learning methods

The heart is an important organ in the human body, and acute myocardial infarction (AMI) is the leadingcause of death in most countries. Researchers are doing a lot of data analysis work to assist doctors in predicting theheart problem. An analysis of the data related to different health problems and its functions can help in predicting thewellness of this organ with a degree of certainty. Our research reported in this paper consists of two main parts. In thefirst part of the paper, we compare different predictive models of hospital mortality for patients with AMI. All resultspresented in this part are based on real data of about 603 patients from a hospital in the Czech Republic and about184 patients from two hospitals in Syria. Although the learned models may be specific to the data, we also draw moregeneral conclusions that we think are generally valid. In the second part of the paper, because the data is incomplete andimbalanced we develop the Chow–Liu and tree-augmented naive Bayesian to deal with that data in better conditions,and compare the quality of these algorithms with others.

PDF

___

[1] Murphy KP. Machine Learning: A Probabilistic Perspective. Canada: The MIT Press, 2012.
[2] Duda R, Hart P, Stork DG. Pattern Classification. USA: John Wiley and Sons, 2001.
[3] Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning. USA: Springer, 2009.
[4] Koller D, Friedman N. Probabilistic Graphical Models: Principles and Techniques. Canada: MIT Press, 2009.
[5] Friedman N, Geiger D, Goldszmidt M. Bayesian network classifiers, Machine Learning 1997; 29(2): 131-163.
[6] Onisko A, Druzdzel M, Wasyluk H. A Bayesian network model for diagnosis of liver disorders. In: Proceedings of the Eleventh Conference on Biocybernetics and Biomedical Engineering; Warsaw, Poland; 1999. pp. 842-846.
[7] Blanco R, Inza I, Naga PL. Feature selection in Bayesian classifiers for the prognosis of survival of cirrhotic patients treated with TIPS. Journal of Biomedical Informatics 2005; 38 (05): 507-543.
[8] Heckerman D, Horvitz E, Nathwani B. Toward normative expert systems: Part I. The Pathfinder project. Methods of Information in Medicine 1992; 31: 90-105.
[9] Krumholz HM, Normand SLT, Galusha DH, Mattera JA, Rich AS et al. Risk-Adjustment Models for AMI and HF 30-Day Mortality, Methodology, USA: Harvard Medical School, Department of Health Care Policy, 2007.
[10] Vomlel J, Kruaffaffk H, Tůma P, Přeček J, Hutyra M. Machine learning methods for mortality prediction in patients with ST elevation myocardial infarction. In: Proceedings of The Nineth Workshop on Uncertainty Processing WUPES’12; Czech Republic; 2012. pp. 204-213.
[11] Salman I, Vomlel J. A machine learning method for incomplete and imbalanced medical data. In: Proceedings of the 20th Czech-Japan Seminar on Data Analysis and Decision Making Under Uncertainty; Pardubice, Czech Republic; 2017. pp. 188-195.
[12] Chow C, Liu C. Approximating discrete probability distributions with dependence trees. IEEE Transactions on Information Theory 1968; 14: 462-467.
[13] Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P et al. The WEKA data mining software: An update, ACM Sigkdd Explorations 2009; 11 (1): 10-18.
[14] Quinlan R, Kaufmann M. C4.5: Programs for machine learning, Machine Learning 1993; 29(2): 131-163.
[15] Cessie Sle, Houwelingen JC. Ridge estimators in logistic regression. Applied Statistics 1992; 41(1): 191-201.
[16] Duda RO, Hart PE. Pattern classification and scene analysis. Wiley-Interscience 1973; 30(1): 106-110.
[17] Kohavi R. Scaling up the accuracy of Naive-Bayes classifiers: A decision-tree hybrid. In: Proceedings of Second International Conference on Knowledge Discovery and Data Mining; Portland, Oregon, USA; 1996. pp. 202-207.
[18] Cooper GF. A Bayesian method for the induction of probabilistic networks from data, Machine Learning 1992; 9(4): 309-347.
[19] Cohen Ira, Cozman FG, Sebe N, Marcelo C, Huang TS. Semi-supervised learning of classifiers: Theory, algorithms and their application to human-computer interaction. IEEE-Transactions on Pattern Analysis and Machine Intelligence 2004; 26(12): 1553-1568.
[20] Francois OCH, Leray P. Learning the tree augmented Naive Bayes classifier from incomplete datasets. In: Third European Workshop on Probabilistic Graphical Models (PGM); Prague, Czech Republic; 2006. pp. 91-98.
[21] Chawla N, Bowyer K, Hall L, Kegelmeyer W. Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research 2002; 11(16): 321-357