Advertisement

Enhanced Neural Networks and Medical Imaging

  • Luis F. Mingo
  • Fernando Arroyo
  • Carmen Luengo
  • Juan Castellanos
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1689)

Abstract

This paper shows that the application of Enhanced Neural Networks when dealing with classification problems is more powerfull than classical Multilayer Perceptrons. These enhanced networks are able to approximate any function f(χ) using a n-degree polinomial defined by the weights in the connections. Also, the addition of hidden layers in the neural architecture, increases the degree of the output equation associated to output units. So, surfaces generated by these networks are really complex and theoretically they could classify any pattern set with a number n of hidden layers. Results concerning medical imaging, breast cancer diagnosis, are studied along the paper. The proposed architecture improves obtained results using classical networks, due to the implicit data transformation computed as part of the neural architecture.

Keywords

Neural Networks Breast Cancer Approximation Theory 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bottou, L. & Gallinari, P.: A Framework for the Cooperation of Learning Algorithms. Laboratoire de Recherche en Informatique. Université de Paris XI. 91405 Orsay Cedex. France.Google Scholar
  2. 2.
    Bottou, L.; Cortes C. & Vapnik V.: On the Efeectiveness VC Dimension. AT&T Bell Laboratories, Holmdel NJ 07733. USA. (1994).Google Scholar
  3. 3.
    Funahashi, K: On the Approximate REalization of Continuous Mappings by Neural Networks. Neural Networks, Vol. 2. Pp. 183–192. (1989).CrossRefGoogle Scholar
  4. 4.
    Hecht-Nielsen, R.: Kolmogorovś Mapping Neural network Existence Theorem. IEEE First International Conference on Neural Networks. San Diego. Vol III. Pp. 11–14. (1987).Google Scholar
  5. 5.
    Hornik, K.; Stinchombe, M. & White, H.: Universal Approximation of an Unknown Mapping and its Derivates Using Multilayer Feedforwar Networks. Neural Networks, Vol. 3. Pp. 625–633. (1990).CrossRefGoogle Scholar
  6. 6.
    Hornik, K.: Approximation Capabilities of Multilayer Feedforward Networks. Neural Networks, Vol. 2. Pp. 551–560. (1991).Google Scholar
  7. 7.
    Jacobs, R. A.; Jordan M. I. & Barto A. G.: Task Decomposition through Competition in a Modular Connectionist Architecture: The what and where vision tasks. MIT. COINS Technical Paper 90-27. (1990).Google Scholar
  8. 8.
    Jordan M. I. & Rumelhart D. E.: Forward Models: Supervised Learning with a Distal Teacher. MIT Center for Cognitive Science. Occasional Paper #40.Google Scholar
  9. 9.
    Mingo, L.F.; Castellanos, J.; Giménez, V. & Villarrasa, A.: Hierarchical Neural Network System to Breast Cancer Diagnosis. 9th. European Signal Processing Conference. EUSIPCO-98. Signal Procesing IX. Theories and Applications. Rhodes, Greece, 8—11 September. Pp. 2457–2460. (1998).Google Scholar
  10. 10.
    Mingo, L.F.; Arroyo, F.; Luengo, C. & Castellanos J.: Learning Hyper-Surfaces iwth Neural Networks. 11th Scandinavian Conference on Image Analysis. SCIA'99. Kangerlussuaq, Greenland. June 7—11. (1999). SUBMITTEDGoogle Scholar
  11. 11.
    Pican, N.: An Orthogonal Delta Weight Estimator for MLP Architectures. ICNN'96 Proceedings, Washington DC. June 2—6. (1996).Google Scholar
  12. 12.
    Stinchombe, M. & White, H: Multilayer Feedforward Networks are Universal Approximators. Neural Networks, Vol. 2. Pp. 551–560. (1989).Google Scholar
  13. 13.
    Wieland, A. & Leighton, R.: Geometric Analysis of Neural Network Capabilities. IEEE First International conference on Neural Networks, Vol. 3. Pp. 385–392. (1987).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Luis F. Mingo
    • 1
  • Fernando Arroyo
    • 1
  • Carmen Luengo
    • 1
  • Juan Castellanos
    • 2
  1. 1.Dept. de Lenguajes Proyectos y Sistemas InformáticosE.U. de Informática Universidad Politécnica de MadridMadridSpain
  2. 2.Dept. de Inteligencia Artificial, Facultad de InformáticaUniversidad Politécnica de MadridBoadilla del Monte, MadridSpain

Personalised recommendations