Detection of Microcalcifications in Mammograms by the Combination of a Neural Detector and Multiscale Feature Enhancement

  • Diego Andina
  • Antonio Vega-Corona
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2085)


We propose a two steps method for the automatic classifi- cation of microcalcifications in Mammograms. The first step performs the improvement of the visualization of any abnormal lesion through feature enhancement based in multiscale wavelet representations of the mammographic images. In a second step the automatic recognition of microcalcifications is achieved by the application of a Neural Network optimized in the Neyman-Pearson sense. That means that the Neural Network presents a controlled and very low probability of classifying abnormal images as normal.


False Alarm Probability Edge Curve Mammographic Image Abnormal Mammogram Dyadic Wavelet 
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.
    Feig, S.A., Hendrick, R.E.: Risk, Benefit and Controversy in Mammographic Screening. In A Categorical Course in Physics. A.G. Haus and M.J. Yaffe. (eds). Radiological Society of North America. (1993) 119–135Google Scholar
  2. 2.
    Feig, S.A.: Mammogramphic Evaluation of Calcifications. In A Categorical Course in Breast Imaging. (eds). Radiological Society of North America. (1995) 93–105Google Scholar
  3. 3.
    Bassett, L.W.: Mammographic Analysis of Calcifications. Radiologic Clinics of North America. 30 (1992) 93–105Google Scholar
  4. 4.
    Strickland, R.N., Hahn, H.I.: Wavelet Transform for Detecting Microcalcifications in Mammograms. IEEE Trans. on Medical Imaging, Vol. 15. 2 (1996) 218–229Google Scholar
  5. 5.
    Jeine, J., Deans, S.R.: Multirresolution Statistical Analysis of High-Resolution Digital Mammograms. IEEE Trans. on Medical Imaging. Vol. 16. 5 (1997) 503–515Google Scholar
  6. 6.
    Redner, R., Walker, H.: Mixture Densities, Maximum Likelihood and the EM Elgorithm. SIAM Review. 26 (1994) 195–239Google Scholar
  7. 7.
    Mallat, S., Zhong, S.: Characterization of Signals from Multiscale Edge. IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 14. 7 (1992) 710–732Google Scholar
  8. 8.
    Sanz-González, J.L., Andina, D.: Performance Analysis of Neural Network Detectors by Importance Sampling Techniques. Neural Processing Letters. Kluwer Academic Publishers. Netherlands. ISSN 1370-4621, Vol.9. 3 (1999) 257–269Google Scholar
  9. 9.
    Feig, S.A., Shaber, G.S., Patchefsky, A.: Analysis of Clinically Occult and Mammographically Occult Breast Tumors. American Journal of Radiology, Vol. 128. (1997) 403–408Google Scholar
  10. 10.
    Meyer, Y.: Onddelettes et Operateurs. Herman, New York. (1990)Google Scholar
  11. 11.
    Mu, Y., Giger M.L., Doi, K.: Artificial Network in Mammography. Application to Decision Making in the Diagnosis of Breast Cancer. Radiology, Vol. 187. (1993) 81–87Google Scholar
  12. 12.
    Qian, W., Clarke, L.P., Kallergi, M., Clark, R.: Tree-Structured Nonlinear Filters in Digital Mammography. IEEE Trans. on Medical Imaging, Vol. 13. 1 (1994) 25–36Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Diego Andina
    • 1
  • Antonio Vega-Corona
    • 2
  1. 1.Departamento de Señales, Sistemas y Radiocomunicaciones, E.T.S.I. TelecomunicaciónUniversidad Politécnica de MadridSpain
  2. 2.F.I.M.E.EUniversidad de GuanajuatoGuanajuatoMexico

Personalised recommendations