Bayesian Model Combination and Its Application to Cervical Cancer Detection

  • Miriam Martínez
  • Luis Enrique Sucar
  • Hector Gabriel Acosta
  • Nicandro Cruz
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4140)


We have developed a novel methodology to combine several models using a Bayesian approach. The method selects the most relevant attributes from several models, and produces a Bayesian classifier which has a higher classification rate than any of them, and at the same time is very efficient. Based on conditional information measures, the method eliminates irrelevant variables, and joins or eliminates dependent variables; until an optimal Bayesian classifier is obtained. We have applied this method for diagnosis of precursor lesions of cervical cancer. The temporal evolution of the color changes in a sequence of colposcopy images is analyzed, and the resulting curve is fit to an approximate model. In previous work we develop 3 different mathematical models to describe the temporal evolution of each image region, and based on each model to detect regions that could have cancer. In this paper we combine the three models using our methodology and show very high accurracy for cancer detection, superior to any of the 3 original models.


Cervical Cancer Bayesian Network Continuous Attribute Minimum Description Length Dependent Attribute 
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.
    Acosta-Mesa, H., Zitová, B., Ríos-Figueroa, H., Cruz-Ramírez, N., Marín- Hernández, A., Hernández-Jiménez, R., Cocotle-Ronzón, B., Hernández-Galicia, E.: Cervical cancer detection using colposcopic images: a temporal approach. In: Proc. Sixth International Conference on Computer Science (ENC 2005), pp. 158–164. IEEE, Los Alamitos (2005)CrossRefGoogle Scholar
  2. 2.
    Anderson, M., Jordan, A.M., Sharp, F.: A Text and Atlas of Integrated Colposcopy, Mosby (1993)Google Scholar
  3. 3.
    Pogue, W., Zelenchuk, A., Harper, W., Burke, G.C., Burke, E.E., Harper, D.M.: Analysis of acetic acid-induced whitening of high-grade squamous intraepithelial lesions. Journal of Biomedical Optics, 397–403 (2001)Google Scholar
  4. 4.
    Balas, C.: A novel optical imaging method for the early detection, quantitative grading, and mapping of cancerous and precancerous lesions of cervix. IEEE Transactions on medical imaging, 96–104 (2001)Google Scholar
  5. 5.
    Burghardt, E., Hellmuth, P., Girardi, F.: Primary care colposcopy. Thieme (2004)Google Scholar
  6. 6.
    Craine, B.L., O’Toole, C.J., Ji, Q.: Digital imaging colposcopy: Corrected area measurements using shape-from-shading. IEEE Transactions on medical imaging, 1003–1010 (1998)Google Scholar
  7. 7.
    Freund, R., Schapire, E.: A short introduction to boosting. Journal of the Japanese Society for Artificial Intelligence 14, 771–780 (1999)Google Scholar
  8. 8.
    Friedman, N., Geiger, D., Goldszmidt, M.: Bayesian network classifiers. Machine Learning 29, 131–163 (1997)MATHCrossRefGoogle Scholar
  9. 9.
    Ji, Q., Craine, E.: Texture analysis for classification of cervix lesions. IEEE Transactions on medical imaging, 1144–1149 (2000)Google Scholar
  10. 10.
    Martínez, M., Sucar, L.E.: Learning an optimal naive bayes classifier. In: Proc. Congreso de Investigación y Extensión. ITESM (2006)Google Scholar
  11. 11.
    Pazzani, M.J.: Searching for atribute dependencies in bayesian classifiers. Preliminary Papers of the Intelligence and Statistics, 424–429 (1996)Google Scholar
  12. 12.
    Pearl, J.: Probabilistic reasoning in intelligent systems. Morgan Kaufmann, Palo Alto (1988)Google Scholar
  13. 13.
    Sucar, L.E., Gillies, D.F., Gillies, D.A.: Objective probabilities in expert systems. Artificial Intelligence 61, 187–208 (1993)CrossRefMathSciNetGoogle Scholar
  14. 14.
    Vlastos, A.T., Zuluaga, A., Follen, M.: New approaches to cervical cancer screening. Contemporany Ob/Gyn, 87–103 (2002)Google Scholar
  15. 15.
    Yi, X., Kuo, Z., Zhang, C.: Classifier combination based on active learning. In: Proc. of the 17th International Conference on Pattern Recognition (ICPR 2004), pp. 184–187. IEEE, Los Alamitos (2004)Google Scholar
  16. 16.
    Zitova, B.: Image registration methods: a survey. Image and Vision Computing, 977–1000 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Miriam Martínez
    • 1
  • Luis Enrique Sucar
    • 2
  • Hector Gabriel Acosta
    • 3
  • Nicandro Cruz
    • 3
  1. 1.Tecnológico de AcapulcoAcapulco, GuerreroMéxico
  2. 2.INAOETonantzintla, PueblaMéxico
  3. 3.Universidad VeracruzanaXalapa, VeracruzMéxico

Personalised recommendations