A consistency contribution based bayesian network model for medical diagnosis

Yan-Ping Yang
DOI: 10.4236/jbise.2010.35068                

Abstract

This paper presents an effective Bayesian network model for medical diagnosis. The proposed approach consists of two stages. In the first stage, a novel feature selection algorithm with consideration of feature interaction is used to get an undirected network to construct the skeleton of BN as small as possible. In the second stage for greedy search, several methods are integrated together to enhance searching performance by either pruning search space or overcoming the optima of search algorithm. In the experiments, six disease datasets from UCI machine learning database were chosen and six off-the-shelf classification algorithms were used for comparison. The result showed that the proposed approach has better classification accuracy and AUC. The proposed method was also applied in a real world case for hypertension prediction. And it presented good capability of finding high risk factors for hypertension, which is useful for the prevention and treatment of hypertension. Compared with other methods, the proposed method has the better performance.

Keywords

Bayesian Network; Medical Diagnosis;

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	Mitchell, T.M. (1997) Machine learning. McGraw-Hill, New York.
[2]	Jordan, M.I. (1995) Why the logistic function? A tutorial discussion on probabilities and neural networks techno- logy. MIT Computational Cognitive Science Report 9503, Massachusetts.
[3]	Aha, D., Kibler, D. and Albert, M. (1991) Instance-based learning algorithms. Machine Learning, 6(1), 37-66.
[4]	Quinlan, J.R. (1993) Programs for machine learning. San Mateo, Morgan Kaufmann, California.
[5]	Lisboa, P.J.G., Ifeachor, E.C. and Szczepaniak, P.S. (2000) Artificial neural networks in biomedicine. Springer-Ver- lag, London.
[6]	Cooper, G.F. and Herskovits, E. (1992) A Bayesian me- thod for the induction of probabilistic networks from data. Machine Learning, 9(4), 309-347.
[7]	Lucas, P. (2001) Bayesian networks in medicine: A model-based approach to medical decision making. Proceedings of the EUNITE Workshop on Intelligent Systems in Patient Care, Vienna, 73-97.
[8]	Antal, P., Verrelst, H., Timmerman, D., Van Huffel, S., De Moor, B., Vergote, I. and Moreau, Y. (2000) Bayesian networks in ovarian cancer diagnosis: Potentials and limitations. 13th IEEE Symposium on Computer-Based Medical Systems, Texas Medical Center, Houston, 103- 108.
[9]	van Gerven, M., Jurgelenaite, R., Taal, R., Heskes, T. and Lucas, P. (2007) Predicting carcinoid heart disease with the noisy-threshold classifier. Artificial Intelligence in Medicine, 40(1), 45-55.
[10]	Pearl, J. (1988) Probabilistic reasoning in intelligent systems: Networks of plausible inference. 2nd Edition, Morgan Kaufmann, San Francisco.
[11]	Spirtes, P., Glymour, C. and Scheines, R. (2000) Causa- tion, prediction, and search. 2nd Edition, the MIT Press, Cambridge.
[12]	Pradhan, M., Henrion, M., Provan, G., Favero, B.D. and Huang, K. (1996) The sensitivity of belief networks to imprecise probabilities: An experimental investigation. The Artificial Intelligence Journal, 85(1-2), 363-397.
[13]	Cheng, J., et al. (2002) Learning Bayesian networks from data: An information theory based approach. The Artificial Intelligence Journal, 137(1-2), 43-90.
[14]	Cooper, G. and Herskovits, E. (1992) A Bayesian method for the induction of probabilistic networks form data. Machine Learning, 9(4), 309-347.
[15]	Heckerman, D. (1998) A tutorial on learning with Bayesian networks, learning in graphical models. Kluwer Academic Publishers, Dordrecht, 301-354.
[16]	Chickering, D. (1996) Learning Bayesian networks is NP-complete. In: Fisher, D. and Lenz, H., Eds., Learning from Data: Artificial Intelligence and Statistics, 4(1) 121-130.
[17]	Friedman, N., Nachman, I. and Peer, D. (1999) Learning Bayesian network structure from massive datasets: The ‘sparse candidate’ algorithm. Uncertainty in Artificial Intelligence, 15(3), 206-215.
[18]	Tsamardinos, I., Brown, L. and Aliferis, C. (2006) The max-min hill-climbing Bayesian network structure learn- ing algorithm. Machine Learning, 65(1), 31-78.
[19]	Meinshausen, N. and Buhlmann, P. (2006) High dimen- sional graphs and variable selection with the lasso. The Annals of Statistics, 34(3), 1436-1462.
[20]	Heckerman, D., Geiger, D. and Chickering, D. (1995) “Learning Bayesian networks: The combination of knowledge and statistical data. Machine Learning, 20(3), 197-243.
[21]	Tsamardinos, I. and Aliferis, C. (2003) Towards prince- pled feature selection: Relevancy, filters and wrappers. The 9th International Workshop on Artificial Intelligence and Statistics, Florida, 334-342.
[22]	Guyon, I. and Elisseeff, A. (2003) An introduction to variable and feature selection. Journal of Machine Lear- ning Research, 3(7-8), 1157-1182.
[23]	Dash, M. and Liu, H. (2003) Consistency-based search in feature selection. Artificial Intelligence, 151(1-2), 155- 176.
[24]	Hall, M. (1999) Correlation based feature selection for machine learning. Ph.D. Dissertation, University of Waikato, New Zealand.
[25]	Kononenko, I. (1994) Estimating attributes: Analysis and extension of RELIEF. European Conference on Machine Learning, Catania, 171-182.
[26]	Jakulin, A. and Bratko, I. (2003) Analyzing attribute dependencies. PKDD, Ljubljana.
[27]	Jakulin, A. and Bratko, I. (2004) Testing the signifi- cance of attributes interactions. International Conference on Machine Learning, 20, 69-73.
[28]	Zhao, Z. and Liu, H. (2007) Searching for interacting features. Proceedings of International Joint Conference on Artificial Intelligence, Nagoya.
[29]	Blake, C., Keogh, E. and Merz, C.J. (1998) UCI repository of machine learning databases. University of California, Irvine. http://www.ics.uci.edu/~mlearn/MLRepo sitory.html
[30]	Aha, D.W., Kibler, D. and Albert, M.K. (1991) Instance- based learning algorithms. Machine Learning, 6(1), 37-66.
[31]	Witten, I.H. and Frank, E. (2005) Data mining-pracitcal machine learning tools and techniques with JAVA implementations. 2nd Edition, Morgan Kaufmann Publi- shers, California.
[32]	Bouckaert, R.R. (2008) Bayesian network classifiers in weka for version 3-5-7. Artificial Intelligence Tools, 11(3), 369-387
[33]	Cortes, C. and Mohri, M. (2003) AUC optimization vs. error rate Minimization. Advances in Neural Information Processing Systems, 11(10), 356-360.
[34]	Chobanian, A.V., Bakris, G.L., Black, H.R., Cushman, W.C., Green L.A. and Izzo, J.L. (2003) The seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure. The JNC 7 Report, Journal of the American Medical Association, 289(19), 2560-2572.
[35]	(2001-2002) Cardiovascular diseases – Prevention and control. WHO Chemical Vapor Deposition Strategy Conference, Nottingham.
[36]	Kearney, P.M., Whelton, M., Reynolds, K., et al. (2005) Global burden of hypertension: Analysis of worldwide data. Lancet, 365(9455), 217-223.