Advertisement

Value of Digital Image Analysis in Research and Diagnosis of Urine Bladder Cancer

  • A. Dulewicz
  • D. Pietka
  • P. Jaszczak
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 45)

Abstract

In this paper two methods of detecting neoplastic changes in urinary bladder are proposed: a noninvasive method mainly for screening investigations based on image analysis of specimens obtained from urine and quick little invasive method to be used as a support for clinical diagnosis based on material obtained by bladder washing technique. All together 147 cases were analyzed with the help of digital image processing system designed by the authors. Multistage classifiers were constructed to identify positive and negative cases. The results of this study yielded a 71% correct classification rate in the control group, while a 100% rate was obtained among the cancer patients for the noninvasive method. The positive test predictive value was equal to 0,77 and negative test predictive value was equal to 1. For the quick (little invasive method) the results were: a 68% correct classification rate in the control group, while a 86% rate was obtained among the cancer patients. The positive test predictive value was equal to 0,82 and negative test predictive value was equal to 0,73. The results show that both methods could be very promising additional diagnostic tools in detection of urinary bladder cancer.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dulewicz A, Pietka D, Jaszczak P, Nechay A, Sawicki W, Pykao R, Kozmiska E, Borkowski A, (2001) Computer identification of neoplastic urothelial nuclei from the bladder. In Analytical and Quantitative Cytology and Histology, vol. 23,(5), October, 321–329.Google Scholar
  2. 2.
    Boon M.E, Drijver J.S, (1986). Routine cytological staining techniques. Theoretical Background and Practice. Macmillan Education Ltd., London.Google Scholar
  3. 3.
    Dulewicz A, Pietka D, Jaszczak P (2005) Trial of practical computer analysis of urothelial nuclei for cancer detection. In: Progress in bladder cancer research, Nova Biomedical Books, New York.Google Scholar
  4. 4.
    Kurzynski M (1997) Rozpoznawanie obrazw. Oficyna Wydawnicza Politechniki Wrocawskiej. Wroclaw.Google Scholar
  5. 5.
    Russ J.C (1995) The Image Processing handbook. CRC Press, Boca Raton, Ann Arbor, London, TokyoGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • A. Dulewicz
    • 1
  • D. Pietka
    • 1
  • P. Jaszczak
    • 1
  1. 1.Institute of Biocybernetics and Biomedical Engineering PASWarsawPoland

Personalised recommendations