Using the Gamma Test in the Analysis of Classification Models for Time-Series Events in Urodynamics Investigations

  • Steve Hogan
  • Paul Jarvis
  • Ian Wilson
Conference paper


Urodynamics is a clinical test in which time series data is recorded measuring internal pressure readings as the bladder is filled and emptied. Two sets of descriptive statistics based on various pressure events from urodynamics tests have been derived from time series data. The suitability of these statistics for use as inputs for event classification through neural networks is investigated by means of the gamma test. BFGS neural network models are constructed and their classification accuracy measured. Through a comparison of the results, it is shown that the gamma test can be used to predict the reliability of models before the neural network training phase begins.


Hide Layer Mean Square Error Classification Accuracy Classification Model Pressure Event 
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.
    Abrams, P., Urodynamics. 2006: Springer Verlag.Google Scholar
  2. 2.
    Evans, D. and A.J. Jones, A proof of the Gamma test. Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 2002. 458(2027): p. 2759.MathSciNetMATHCrossRefGoogle Scholar
  3. 3.
    Broyden, C., et al., BFGS method. Journal of the Institute of Mathematics and Its Applications, 1970. 6: p. 76-90.MathSciNetMATHCrossRefGoogle Scholar
  4. 4.
    Zurada, J.M., Introduction to artificial neural systems. 1992.Google Scholar
  5. 5.
    Stefánsson, A., N. Končar, and A.J. Jones, A note on the Gamma test. Neural Computing & Applications, 1997. 5(3): p. 131-133.CrossRefGoogle Scholar
  6. 6.
    Kemp, S., I. Wilson, and J. Ware, A tutorial on the gamma test. International Journal of Simulation: Systems, Science and Technology, 2004. 6(1-2): p. 67–75.Google Scholar
  7. 7.
    Evans, D., Data-derived estimates of noise for known smooth models using near-neighbour asymptotics, in Department of Computer Science. 2002, Cardiff University Cardiff.Google Scholar
  8. 8.
    Dreiseitl, S. and L. Ohno-Machado, Logistic regression and artificial neural network classification models: a methodology review. Journal of Biomedical Informatics, 2002. 35(5-6): p. 352-359.CrossRefGoogle Scholar
  9. 9.
    Baxt, W. G., Application of artificial neural networks to clinical medicine. The lancet, 1995. 346(8983): p. 1135-1138.CrossRefGoogle Scholar
  10. 10.
    Chai, S. S., et al., Backpropagation neural network for soil moisture retrieval using NAFE’05 data: a comparison of different training algorithms. Int Archives Photogramm, Remote Sens Spatial Inf Sci (China), 2008. 37: p. 1345.Google Scholar
  11. 11.
    Xia, J.H. and A. Kumta, Feedforward Neural Network Trained by BFGS Algorithm for Modeling Plasma Etching of Silicon Carbide. Plasma Science, IEEE Transactions on, 2010. 38(2): p. 142-148.CrossRefGoogle Scholar
  12. 12.
    Chester, D.L. Why two hidden layers are better than one. 1990.Google Scholar
  13. 13.
    Cybenko, G., Continuous valued neural networks with two hidden layers are sufficient. Mathematics of Control, Signal and Systems, 1989. 2: p. 303-314.MathSciNetMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2011

Authors and Affiliations

  1. 1.University of GlamorganPontypriddUK

Personalised recommendations