Advertisement

Quality assurance and increased efficiency in medical projects with neural networks by using a structured development method for feedforward neural networks (SENN)

  • T. Waschulzik
  • W. Brauer
  • M. Förster
  • K. Kirchner
  • R. Engelbrecht
  • T. Schütz
  • T. Koschinsky
  • G. Entenmann
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 934)

Abstract

The growing number of projects using neural networks in medical care makes it necessary to examine how productivity can be increased and how quality can be assured. This examination addresses the problem specification, data preparation as well as the development of appropriate representations, the selection of suitable encodings and the combination of encodings. Network paradigms with fast learning properties and network structures that can be analysed and interpreted after the training process have been successfully applied to medical tasks. The associative recall of examples (ARE) can be used to verify the quality of representations and encodings. Furthermore it is possible to evaluate the competence of a neural network for a specific task by an ARE. Finally, a standard approach and the hereafter presented method applied for a medical project are compared. The comparison of these two approaches and the collection of the medical data is part of the DIADOQ-project.

Keywords

Neural Network Secondary Failure Medical Project System Life Cycle Spiral 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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anderson J.A., Rosenfeld E.: Neurocomputing — Foundations of Research MIT Press (1988)Google Scholar
  2. Andreassen S., Engelbrecht R., Wyatt J. (eds), Technology and Informatics 10, Artificial Intelligence in Medicine, p 466–476. IOS Press, Amsterdam (1993)Google Scholar
  3. Benaim M.: On functional Approximation with Normalized Gaussian Units Neural Computation 6, p 319–333 (1994)Google Scholar
  4. Boehm B.W.:A spiral model of software development and enhancement. Computer, May 1988, pp.61–72Google Scholar
  5. Eldracher M., Waschulzik T.: A Topologically Distributed Encoding to Facilitate Learning. Journal of Systems Engineering 3, p 110–119. Springer, London (1993)Google Scholar
  6. Egmont-Peterson M., Talmon, J. L., Brender J., NcNair P.: On the quality of neural net classifiers. Artificial Intelligence in Medicine, Vol. 6, No.5, pp. 359–381,1994PubMedGoogle Scholar
  7. Engelbrecht R., Rector A., Moser W: Verification and validation. In: Assessment and Evaluation of Information Technologies in Medicine (Eds.: van Gennip E.M.S.J.,Talmon J.L.) Amsterdam, IOS Press, p 51–66 (1994)Google Scholar
  8. European NIDDM Policy Group: A Desktop Guide for the Management of non-insulin-dependent diabetes mellitus. IDF Bulletin 25, 1990 1 (1990)Google Scholar
  9. Fahlman S. E., Lebiere C: The Cascade-Correlation Learning Architecture. School of Computer Science, Carnegie Mellon University, Pittsburg, USA (1991)Google Scholar
  10. Förster M, Kirchner K, Waschulzik T, et al.: DIADOQ: “Untersuchungen zum Aufbau von Wissensbasen in der Diabetologie mittels Neuronaler Netze”. 39. Jahrestagung der GMDS, Dresden (1994) (in press)Google Scholar
  11. Geiger H.:Storing and Processing Information in Connectionist Systems.-In: R. Eckmiller, editor, Advanced Neural Computers, p. 271–277. Elsevier Science Publishers B.V. (North Holland) (1990)Google Scholar
  12. Hinton G.E. McClelland J. L., Rummelhart D. E.: Distributed Representations p77–109. In: Rummelhart D. E., McClelland J.L. and the PDP Research Group: Parallel Distributed Processing, MIT Press (1986)Google Scholar
  13. Ifeachor E., Rosen K. (eds). Proceedings of the International Conference on Neural Networks and Expert Systems in Medicine and Healthcare, Plymouth (1994)Google Scholar
  14. Minsky M., Papert S.: Perceptrons, Cambridge, MA: MIT-Press (1969)Google Scholar
  15. Nelson M. M., Inllingwonh W.T.: A practical Guide to Neural Nets, Addison-Wesley Publishing Company (1990)Google Scholar
  16. Nöhmeier M., Waschulzik T., Brauer W., Grothe F., Engelbrecht R.: A Rule-based System for Realisation of Medical Projects with Neural Networks (HEAD). In: Ifeachor E., Rosen K. (eds). Proceedings of the International Conference on Neural Networks and Expert Systems in Medicine and Healthcare, Plymouth (1994)Google Scholar
  17. Park J., Sandberg I. W.: Universal approximation using radial-basis-function networks. Neural Computation 3, p 246–257 (1991)Google Scholar
  18. Polzer A.: Invariante Mustererkennung durch neuronale Netzwerke mit großen rezeptiven Feldern.-Diploma thesis, München. 1992Google Scholar
  19. Rosenblatt F.: The perceptron: a probabilistic model for information storage and organization in the brain. Physiological Review 65, p 386–408 (1958)Google Scholar
  20. Rumelhart D. E., Hinton G. E., Williams R. J.: Learning representations by back-propagating errors. Nature 323, p 533–536 (1986)CrossRefGoogle Scholar
  21. Waschulzik T., Geiger H.: Eine Entwicklungsmethodik für strukturierte konnektionistische Systeme.-In: G. Dorffner (eds), Informatik Fachberichte 252, Konnektionismus in Artificial Intelligence und Kognitionsforschung, p. 202–206. Springer, Berlin (1990a)Google Scholar
  22. Werbos P. J.: Beyond regression: new tools for prediction and analysis in the behavioral sciences. Ph. D. thesis. Harvard University, Cambridge, MA (1974)Google Scholar
  23. Widrow B., Hoff M. E.: Adaptive switching circuits. 1960 IRE WESCON Convention Record, New York: IRE, p 96–104 (1960)Google Scholar
  24. Yki-Järvinnen H.:Pathogenesis of non-Insulin-dcpendent diabetes mellitus. The Lancet Vol 343, p 91–95 (1994)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • T. Waschulzik
    • 1
  • W. Brauer
    • 2
  • M. Förster
    • 3
  • K. Kirchner
    • 4
  • R. Engelbrecht
    • 1
  • T. Schütz
    • 1
  • T. Koschinsky
    • 4
  • G. Entenmann
    • 4
  1. 1.GSF-MEDIS-InstitutOberschleißheimGermany
  2. 2.Institut für InformatikTechnische Universität MünchenMünchen 2Germany
  3. 3.Institut für Medizinische Informatik und. BiometrieDresdenGermany
  4. 4.Diabetes ForschungsinstitutDüsseldorfGermany

Personalised recommendations