An Improved EM Algorithm for Statistical Segmentation of Brain MRI

  • Yong Yang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4224)


To overcome the limitations of standard expectation maximization (EM) algorithm, an improved EM algorithm is proposed. Based on this algorithm, a novel statistical approach for segmentation of brain magnetic resonance (MR) image data is presented in this paper, which involves three steps. Firstly, after pre-processing the image with the curvature anisotropic diffusion filter, the background (BG) and brain masks of the image are obtained by applying a combination approach of thresholding with morphology. Secondly, the connected threshold region growing technique is employed to get the preliminary results of white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) on a brain MRI. Finally, the previous results are served as the priori knowledge for the improved EM algorithm to segment the brain MRI. The performance of the proposed method is compared with those of standard EM algorithm and the popular used fuzzy-C means (FCM) segmentation. Experimental results show our approach is effective, robust and significantly faster than the conventional EM based method.


Gray Matter Expectation Maximization Brain Magnetic Resonance Imaging Expectation Maximization Algorithm Statistical Segmentation 
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. Wells, W.M., Grimson, W.E.L., Kikins, R., et al.: Adaptive segmentation of MRI data. IEEE Trans. Med. Imag. 15, 429–442 (1996)CrossRefGoogle Scholar
  2. Held, K., Kops, E.R., Krause, B.J., et al.: Markov random field segmentation of brain MR images. IEEE Trans. Med. Imag. 16, 878–886 (1997)CrossRefGoogle Scholar
  3. Atkins, M.S., Mackiewich, B.T.: Fully automatic segmentation of the brain in MRI. IEEE Trans. Med. Imag. 17, 98–107 (1998)CrossRefGoogle Scholar
  4. Clarke, L.P., Velthuizen, R.P., Camacho, M.A., et al.: MRI segmentation: Methods and applications. Magn. Reson. Imag. 13, 343–368 (1995)CrossRefGoogle Scholar
  5. Rajapakse, J.C., Giedd, J.N., Rapoport, J.L.: Statistical approach to segmentation of single-channel cerebral MR images. IEEE Trans. Med. Imag. 16, 176–186 (1997)CrossRefGoogle Scholar
  6. Hashimoto, A., Kudo, H.: Ordered-subsets EM algorithm for image segmentation with application to brain MRI. In: Proc. IEEE Nuclear Science Symposium Conference Record, vol. 3, pp. 118–121 (2000)Google Scholar
  7. Liang, Z., MacFall, J.R., Harrington, D.P.: Parameter estimation and tissue segmentation from multispectral MR images. IEEE Trans. Med. Imag. 13, 441–449 (1994)CrossRefGoogle Scholar
  8. Dempster, P., Laird, N.M., Rubin, D.B.: Maximum-likelihood from incomplete data via the EM algorithm. J. Roy. Statist. Soc. 39, 1–38 (1977)MATHMathSciNetGoogle Scholar
  9. Gerig, G., Kübler, O., Kikinis, R., et al.: Nonlinear anisotropic filtering of MRI data. IEEE Trans. Med. Imag. 11, 221–232 (1992)CrossRefGoogle Scholar
  10. Collins, D.L., Zijdenbos, A.P., Kollokian, V., et al.: Design and construction of a realistic digital brain phantom. IEEE Trans. Med. Imag. 17, 463–468 (1998)CrossRefGoogle Scholar
  11. Bezdek, J.: Pattern Recognition with Fuzzy Objective Functions Algorithms. Plenum Press, New York (1981)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Yong Yang
    • 1
  1. 1.School of Information ManagementJiangxi University of Finance and EconomicsNanchangP.R. China

Personalised recommendations