Automatic Detection of Spiculated Masses Using Fractal Analysis in Digital Mammography

  • HyungJun Kim
  • WonHa Kim
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3691)


This paper describes the development of a system for the automatic detection of spiculated masses in digital mammography. We have adopted the lattice space as an image domain model and a distance measure to describe image shapes precisely in pixel scale. Based on the model, we have developed a method to automatically determine binary mass shapes. The method is robust against noise and background brightness of mass region. We also have proposed a novel method in mapping a mass shape into a one-dimensional profile, and then quantified the irregularity of the mass shape by calculating the fractal dimension of the profile. Preliminary experimental results support the hypothesis of the spiculation detection performance of the proposed method would show a possible solution for finding spiculated masses.


Fractal Dimension Binary Image Fractal Analysis Automatic Detection Digital Mammography 
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.
    Karssemeijer, N.: Detection of masses in mammogram. In: Strickland, R.N. (ed.) Image Processing Techniques for Tumor Detection, pp. 187–212. Marcel Dekker, Inc., New York (2002)Google Scholar
  2. 2.
    Karssemeijer, N., Brake, G.M.: Detection of stellate distortions in mammogram. IEEE Trans. Medical Imaging 15(5), 611–619 (1996)CrossRefGoogle Scholar
  3. 3.
    Brake, G.M., Karssemeijer, N.: Single and multiscale detection of masses in digital mammograms. IEEE Trans. Medical Imaging 18(7), 628–639 (1999)CrossRefGoogle Scholar
  4. 4.
    Huang, S., Chang, R., Chen, D., Moon, W.: Characterization of spiculation on ultrasound lesions. IEEE Trans. Medical Imaging 23(1), 111–121 (2004)CrossRefGoogle Scholar
  5. 5.
    Pohlman, S., Powell, K., Obuchowski, N., Chilcote, W., Grundfest-Broniatowski, S.: Quantitative classification of breast tumors in digitized mammograms. Med. Phys. 23(8), 1337–1345 (1996)CrossRefGoogle Scholar
  6. 6.
    Otsu, N.: A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. SMC-9, 62–66 (1979)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • HyungJun Kim
    • 1
  • WonHa Kim
    • 2
  1. 1.Graduate School of Information SecurityKorea University 
  2. 2.School of Electronics & Information EngineeringKyung Hee University 

Personalised recommendations