An intuitive risk factors search algorithm: usage of the Bayesian network technique in personalized medicine. (English) Zbl 1514.62747
Summary: The article focuses on the application of the Bayesian networks (BN) technique to problems of personalized medicine. The simple (intuitive) algorithm of BN optimization with respect to the number of nodes using naive network topology is developed. This algorithm allows to increase the BN prediction quality and to identify the most important variables of the network. The parallel program implementing the algorithm has demonstrated good scalability with an increase in the computational cores number, and it can be applied to the large patients database containing thousands of variables. This program is applied for the prediction for the unfavorable outcome of coronary artery disease (CAD) for patients who survived the acute coronary syndrome (ACS). As a result, the quality of the predictions of the investigated networks was significantly improved and the most important risk factors were detected. The significance of the tumor necrosis factor-alpha gene polymorphism for the prediction of the unfavorable outcome of CAD for patients survived after ACS was revealed for the first time.
MSC:
| 62-XX | Statistics |
Keywords:
Bayesian networks; variable correlation; personalized medicine; naive network optimization; acute coronary syndrome; TNF gene polymorphismReferences:
| [1] | S. Andreassen, M. Woldbye, B. Falck and S.K. Andersen, MUNIN - A Causal Probabilistic Network for Interpretation of Electromyographic Findings, Proceedings 10th International Joint Conference on Artificial Intelligence (IJCAI), Milan, Italy, 1987, pp. 366-372. |
| [2] | M. Athanasiou and J. Clark, A Bayesian network model for the diagnosis of the caring procedure for wheelchair users with spinal injury, Comput. Methods Programs Biomed. 95 (2009), pp. S44-S54. doi: 10.1016/j.cmpb.2009.02.018 · doi:10.1016/j.cmpb.2009.02.018 |
| [3] | A. Aussema, S.R. de Moraisa, and M. Corbex, Analysis of nasopharyngeal carcinoma risk factors with Bayesian networks, Artif. Intell. Med. 54 (2012), pp. 53-62. doi: 10.1016/j.artmed.2011.09.002 · doi:10.1016/j.artmed.2011.09.002 |
| [4] | D. Barber, Bayesian Reasoning and Machine Learning, Cambridge University Press, Cambridge, 2011, pp. 57-73. · Zbl 1267.68001 · doi:10.1017/CBO9780511804779 |
| [5] | E.S. Burnside, D.L. Rubin, J.P. Fine, R.D. Shachter, G.A. Sisney, and W.K. Leung, Bayesian network to predict breast cancer risk of mammographic microcalcifications and reduce number of benign biopsy results: Initial experience, Radiology 240 (2006), pp. 666-673. doi: 10.1148/radiol.2403051096 · doi:10.1148/radiol.2403051096 |
| [6] | E. Castillo, J.M. Gutiérrez, and A.S. Hadi, Sensitivity analysis in discrete Bayesian networks, IEEE Trans. Syst. 27 (1997), pp. 412-423. · doi:10.1109/3468.594909 |
| [7] | J.P. Choi, T.H. Han, and R.W. Park, A hybrid Bayesian network model for predicting breast cancer prognosis, J. Korean Soc. Med. Inform. 15 (2009), pp. 49-57. doi: 10.4258/jksmi.2009.15.1.49 · doi:10.4258/jksmi.2009.15.1.49 |
| [8] | G. Cooper, Nestor: A computer-based medical diagnosis that integrates causal and probabilistic knowledge, Technical Report HPP-84-48, Stanford University, CA, 1984. |
| [9] | G.F. Cooper and E. Herskovits, A Bayesian method for the induction of probabilistic networks from data, Mach. Learn. 9 (1992), pp. 309-347. · Zbl 0766.68109 · doi:10.1007/BF00994110 |
| [10] | J. Davis, E. Lantz, D. Page, J. Struyf, P. Peissig, H. Vidaillet, and M. Caldwell, Machine Learning for Personalized Medicine: Will This Drug Give Me a Heart Attack? Proceedings of Machine Learning in Health Care Applications Workshop. In conjunction with ICML 2008, July 9, Helsinki, Finland, 2008. |
| [11] | Description of ‘Chebyshev’ supercomputer, Research Computing Center of Moscow State University, Moscow. Available at http://www.parallel.ru/cluster/skif_msu.html. |
| [12] | Description of ‘Lomonosov’ supercomputer, Research Computing Center of Moscow State University, Moscow. Available at http://www.parallel.ru/cluster/lomonosov.html. |
| [13] | P. Domingos and M. Pazzani, Beyond independence: Conditions for the optimality of the simple Bayesian classifier, Proceedings of the 13th International Conference on Machine Learning, Bari, Italy, 1996, pp. 105-112. |
| [14] | T. Fawcett, ROC Graphs: Notes and Practical Considerations for Researchers, Kluwer Academic Publishers, Netherlands, 2004. |
| [15] | U. Fayyad, G. Piatetsky-Shapiro, and P. Smyth, From data mining to knowledge discovery in databases, AI Mag. 17 (1996), pp. 37-54. |
| [16] | J. Friedman, On bias, variance, 0/1 - loss, and the curse-of-dimensionality, Data Min. Knowl. Discov. 1 (1997), pp. 55-77. doi: 10.1023/A:1009778005914 · doi:10.1023/A:1009778005914 |
| [17] | N. Friedman, The Bayesian Structural EM Algorithm, Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence (UAI), 1998, pp. 129-138. |
| [18] | N. Friedman, D. Geiger, and M. Goldszmidt, Bayesian networks classifiers, Mach. Learn. 29 (1997), pp. 131-163. doi: 10.1023/A:1007465528199 · Zbl 0892.68077 · doi:10.1023/A:1007465528199 |
| [19] | A.P. Goncalves, J. Ferreira, C. Aguiar, and R. Seabra-Gomes, TIMI, PURSUIT, and GRACE risk scores: Sustained prognostic value and interaction with revascularization in NSTE-ACS, Eur. Heart J. 26 (2005), pp. 865-872. doi: 10.1093/eurheartj/ehi187 · doi:10.1093/eurheartj/ehi187 |
| [20] | A. Gruber and I. Ben-Gal, Efficient Bayesian network learning for system optimization in reliability engineering, Qual. Technol. Quant. Manag. 9 (2012), pp. 97-114. |
| [21] | J.A. Hanley and B.J. McNeil, The meaning and use of the area under a receiver operating characteristic (ROC) curve, Radiology 143 (1982), pp. 29-36. · doi:10.1148/radiology.143.1.7063747 |
| [22] | T. Hastie, R. Tibshirani and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Springer Science+Business Media, LLC, New York, 2009. · Zbl 1273.62005 · doi:10.1007/978-0-387-84858-7 |
| [23] | D. Heckerman, E. Horwitz, and B. Nathwani, Towards normative expert systems: Part I - the Pathfinder project, Methods Inf. Med. 31 (1992), pp. 90-105. |
| [24] | F.V. Jensen, S.H. Aldenryd, and K.B. Jensen, Sensitivity analysis in Bayesian networks, Lecture Notes Comput. Sci. 946 (1995), pp. 243-250. doi: 10.1007/3-540-60112-0_28 · doi:10.1007/3-540-60112-0_28 |
| [25] | F.V. Jensen and T.D. Nielsen, Bayesian Networks and Decision Graphs, Springer, New York, 2007. · Zbl 1277.62007 · doi:10.1007/978-0-387-68282-2 |
| [26] | J.H. Kim and J. Pearl, CONVINCE: A conversational inference consolidation engine, IEEE Trans. Syst. Man Cybernet. 17 (1987), pp. 120-132. doi: 10.1109/TSMC.1987.4309025 · doi:10.1109/TSMC.1987.4309025 |
| [27] | U.B. Kjærulff, A.L. Madsen, Bayesian Networks and Influence Diagrams, Springer, New York, 2008. · Zbl 1251.68001 · doi:10.1007/978-0-387-74101-7 |
| [28] | G.D. Kleiter, Bayesian diagnosis in expert systems, Art. Int. 54 (1992), pp. 1-32. doi: 10.1016/0004-3702(92)90086-D · Zbl 0780.68106 · doi:10.1016/0004-3702(92)90086-D |
| [29] | K.B. Korb and A.E. Nicholson, Bayesian Artificial Intelligence, Chapman and Hall/CRC, London, 2004. · Zbl 1080.68100 |
| [30] | W. Lam and F. Bacchus, Learning Bayesian belief networks. An approach based on the MDL principle, Comput. Intell. 10 (1994), pp. 269-293. doi: 10.1111/j.1467-8640.1994.tb00166.x · doi:10.1111/j.1467-8640.1994.tb00166.x |
| [31] | R.M. Locksley, N. Killeen, and M.J. Lenardo, The TNF and TNF receptor superfamilies, Cell 104 (2001), pp. 487-501. doi: 10.1016/S0092-8674(01)00237-9 · doi:10.1016/S0092-8674(01)00237-9 |
| [32] | A. Mittal and A. Kassim, Bayesian Network Technologies: Applications and Graphical Models, IGI Publishing, Hershey, New York, 2007. · doi:10.4018/978-1-59904-141-4 |
| [33] | S. Mendis, P. Puska, and B. Norrving (eds.), Global Atlas on Cardiovascular Disease Prevention and Control, WHO Press, Geneva, 2011. |
| [34] | N.A. Obuchowski, Fundamentals of clinical research for radiologists: ROC analysis, AJR 184 (2005), pp. 364-372. doi: 10.2214/ajr.184.2.01840364 · doi:10.2214/ajr.184.2.01840364 |
| [35] | J. Pearl, Probabilistic Reasoning in Intelligent Systems, Morgan Kaufman, San Mateo, CA, 1988. · Zbl 0746.68089 |
| [36] | J.L. Rodgers and W.A. Nicewander, Thirteen ways to look at the correlation coefficient, Amer. Stat. 42 (1988), pp. 59-66. doi: 10.2307/2685263 · doi:10.2307/2685263 |
| [37] | P. Sebastiani, M.F. Ramoni, V. Nolan, C.T. Baldwin, and M.H. Steinberg, Genetic dissection and prognostic modeling of overt stroke in sickle cell anemia, Nature Genet. 37 (2005), pp. 435-440. doi: 10.1038/ng1533 · doi:10.1038/ng1533 |
| [38] | J. Tao, Q. Li, C. Zhu, and J. Li, A hierarchical naive Bayesian network classifier embedded GMM for textural image, Int. J. Appl. Earth Observation Geoinformation 14 (2012), pp. 139-148. doi: 10.1016/j.jag.2011.08.012 · doi:10.1016/j.jag.2011.08.012 |
| [39] | M.A. Valle, S. Varas and G.A. Ruz, Job performance prediction in a call center using a naive Bayes classifier, Expert Syst. Appl. 39 (2012), pp. 9939-9945. doi: 10.1016/j.eswa.2011.11.126 · doi:10.1016/j.eswa.2011.11.126 |
| [40] | H. Volzke, C. Schmidt, S. Baumeister, T. Ittermann, G. Fung, J. Krafczyk-Korth, W. Hoffmann, M. Schwab, H. Meyer zu Schwabedissen, M. Dorr, S. Felix, W. Lieb, and H. Kroemer, Personalized cardiovascular medicine: concepts and methodological considerations, Nat. Rev. Cardiol. 10 (2013), pp. 308-316. doi: 10.1038/nrcardio.2013.35 · doi:10.1038/nrcardio.2013.35 |
| [41] | C. Wang, N. Komodakis, and N. Paragios, Markov Random Field modeling, inference & learning in computer vision & image understanding: A survey, Comput Vision Image Understanding 117 (2013), pp. 1610-1627. doi: 10.1016/j.cviu.2013.07.004 · doi:10.1016/j.cviu.2013.07.004 |
| [42] | X. Wu and V. Kumar, The Top Ten Algorithms in Data Mining, Chapman and Hall/CRC, Taylor & Francis Group, Boca Raton, FL, 2009, pp. 37-59. · Zbl 1179.68129 · doi:10.1201/9781420089653 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
