Ensembles of Bayesian Network Classifiers Using Glaucoma Data and Expertise
- 1.4k Downloads
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.
KeywordsBayesian Network Weighted Vote Conditional Probability Distribution Combine Network Probabilistic Graphical Model
Unable to display preview. Download preview PDF.
- 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.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
- 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.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.Garway-Heath, D.F.: Moorfields regression analysis. The Essential HRT Primer. Jocoto Advertising, San Ramon (2005)Google Scholar
- 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
- 14.Heckerman, D.: A tutorial on learning with Bayesian networks. Tech. Report, Microsoft Research (1995)Google Scholar
- 20.Pearl, J.: Probabilistic reasoning in intelligent systems: Networks of plausible inference. Morgan Kaufmann, San Francisco (1988)Google Scholar
- 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
- 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.Yanoff, M., Duker, J.S.: Ophthalmology. Mosby, St. Louis (2003)Google Scholar