Ensembles of Bayesian Network Classifiers Using Glaucoma Data and Expertise

Part of the Studies in Computational Intelligence book series (SCI, volume 373)


Bayesian Networks (BNs) are probabilistic graphical models that are popular in numerous fields. Here we propose these models to improve the classification of glaucoma, a major cause of blindness worldwide. We use visual field and retinal data to predict the early onset of glaucoma. In particular, the ability of BNs to deal with missing data allows us to select an optimal data-driven network by comparing supervised and semi-supervised models. An expertise-driven BN is also built by encoding expert knowledge in terms of relations between variables. In order to improve the overall performances for classification and to explore the relations between glaucomatous data and expert knowledge, the expertise-driven network is combined with the selected data-driven network using a BN-based approach. An accuracy-weighted combination of these networks is also compared to the other models. The best performances are obtained with the semi-supervised data-driven network. However, combining it with the expertise-driven network improves performance in many cases and leads to interesting insights about the datasets, networks and metrics.


Bayesian Network Weighted Vote Conditional Probability Distribution Combine Network Probabilistic Graphical Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Artes, P.H., Chauhan, B.C.: Longitudinal changes in the visual field and optic disc in glaucoma. Progress in Retinal and Eye Research 24, 333–354 (2005)CrossRefGoogle Scholar
  2. 2.
    Bowd, C., Zangwill, L.M., Medeiros, F.A., Tavares, I.M., Hoffmann, E.M., Bourne, R.R., Sample, P.A., Weinreb, R.N.: Structure-function relationships using confocal scanning laser ophthalmoscopy, optical coherence tomography, and scanning laser polarimetry. Investigative Ophthalmology & Visual Science 47, 2889 (2006)CrossRefGoogle Scholar
  3. 3.
    Chauhan, B.C., Drance, S.M., Douglas, G.R.: The use of visual field indices in detecting changes in the visual field in glaucoma. Investigative Ophthalmology & Visual Science 31, 512 (1990)Google Scholar
  4. 4.
    Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the em algorithm. J. the Royal Stat. Society. Series B (Methodological) 39, 1–38 (1977)MathSciNetzbMATHGoogle Scholar
  5. 5.
    Duin, R., Tax, D.: Experiments with classifier combining rules. In: Kittler, J., Roli, F. (eds.) MCS 2000. LNCS, vol. 1857, pp. 16–29. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  6. 6.
    Efron, B., Tibshirani, R., Tibshirani, R.J.: An introduction to the bootstrap. Chapman & Hall/CRC Press, Boca Raton (1993)zbMATHGoogle Scholar
  7. 7.
    Friedman, N., Koller, D.: Being Bayesian about network structure: A Bayesian approach to structure discovery in Bayesian networks. Machine Learning 50, 95–125 (2003)CrossRefzbMATHGoogle Scholar
  8. 8.
    Gaasterland, D.E., Ederer, F., Sullivan, E.K., Caprioli, J., Cyrlin, M.N.: Advanced glaucoma intervention study: 2. visual field test scoring and reliability. Ophthalmology 101, 1445–1455 (1994)Google Scholar
  9. 9.
    Garg, A., Pavlovic, V., Huang, T.S.: Bayesian networks as ensemble of classifiers. In: Proc. the 16th Int. Conf. Pattern Recogn., Quebec, Canada, pp. 779–784. IEEE Comp. Society, Los Alamitos (2002)Google Scholar
  10. 10.
    Garway-Heath, D.F.: Moorfields regression analysis. The Essential HRT Primer. Jocoto Advertising, San Ramon (2005)Google Scholar
  11. 11.
    Garway-Heath, D.F., Poinoosawmy, D., Fitzke, F.W., Hitchings, R.A.: Mapping the visual field to the optic disc in normal tension glaucoma eyes. Ophthalmology 107, 1809–1815 (2000)CrossRefGoogle Scholar
  12. 12.
    Goldbaum, M.H., Sample, P.A., White, H., Colt, B., Raphaelian, P., Fechtner, R.D., Weinreb, R.N.: Interpretation of automated perimetry for glaucoma by neural network. Investigative Ophthalmology & Visual Science 35, 3362 (1994)Google Scholar
  13. 13.
    Hansen, L.K., Salamon, P.: Neural network ensembles. IEEE Trans. Pattern Analysis and Machine Intell. 12, 993–1001 (1990)CrossRefGoogle Scholar
  14. 14.
    Heckerman, D.: A tutorial on learning with Bayesian networks. Tech. Report, Microsoft Research (1995)Google Scholar
  15. 15.
    Jacobs, R.A., Jordan, M.I., Nowlan, S.J., Hinton, G.E.: Adaptive mixtures of local experts. Neural Comp. 3, 79–87 (1991)CrossRefGoogle Scholar
  16. 16.
    Johnson, C.A., Sample, P.A., Zangwill, L.M., Vasil, C.G., Cioffi, G.A., Liebmann, J.R., Weinreb, R.N.: Structure and function evaluation (SAFE): II. Comparison of optic disk and visual field characteristics. American J. Ophthalmology 135, 148–154 (2003)CrossRefGoogle Scholar
  17. 17.
    Kirkpatrick, S., Gelatt, C.D., Vecchi Jr., M.P.: Optimization by simulated annealing. Science 220, 671 (1983)MathSciNetCrossRefGoogle Scholar
  18. 18.
    Kittler, J.: Combining classifiers: A theoretical framework. Pattern Analysis & Appl. 1, 18–27 (1998)CrossRefGoogle Scholar
  19. 19.
    Madigan, D., Raftery, A.E.: Model selection and accounting for model uncertainty in graphical models using Occam’s window. J. the American Stat. Association 89, 1535–1546 (1994)CrossRefzbMATHGoogle Scholar
  20. 20.
    Pearl, J.: Probabilistic reasoning in intelligent systems: Networks of plausible inference. Morgan Kaufmann, San Francisco (1988)Google Scholar
  21. 21.
    Resnikoff, S., Pascolini, D., Etya’ale, D., Kocur, I., Pararajasegaram, R., Pokharel, G.P., Mariotti, S.P.: Global data on visual impairment in the year 2002. Bulletin of the World Health Organization 82, 844–851 (2004)Google Scholar
  22. 22.
    Sharkey, A.J.C.: On combining artificial neural nets. Connection Science 8, 299–314 (1996)CrossRefGoogle Scholar
  23. 23.
    Tucker, A., Vinciotti, V., Liu, X., Garway-Heath, D.: A spatio-temporal Bayesian network classifier for understanding visual field deterioration. Artif. Intell. Medicine 34, 163–177 (2005)CrossRefGoogle Scholar
  24. 24.
    Valentini, G., Masulli, F.: Ensembles of learning machines. In: Marinaro, M., Tagliaferri, R. (eds.) WIRN 2002. LNCS, vol. 2486, pp. 3–20. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  25. 25.
    Wollstein, G., Garway-Heath, D.F., Hitchings, R.A.: Identification of early glaucoma cases with the scanning laser ophthalmoscope. Ophthalmology 105, 1557–1563 (1998)CrossRefGoogle Scholar
  26. 26.
    Woods, K., Bowyer, K., Kegelmeyer Jr., W.P.: Combination of multiple classifiers using local accuracy estimates. In: Proc. 1996 IEEE Comp. Society Conf. Comp. Vision and Pattern Recogn., San Francisco, CA, pp. 391–396. IEEE Comp. Society, Los Alamitos (1996)Google Scholar
  27. 27.
    Yanoff, M., Duker, J.S.: Ophthalmology. Mosby, St. Louis (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  1. 1.Department of Information Systems and ComputingBrunel UniversityLondonUK

Personalised recommendations