Objective PET Lesion Segmentation Using a Spherical Mean Shift Algorithm

  • Thomas B. Sebastian
  • Ravindra M. Manjeshwar
  • Timothy J. Akhurst
  • James V. Miller
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4191)


PET imagery is a valuable oncology tool for characterizing lesions and assessing lesion response to therapy. These assessments require accurate delineation of the lesion. This is a challenging task for clinicians due to small tumor sizes, blurred boundaries from the large point-spread-function and respiratory motion, inhomogeneous uptake, and nearby high uptake regions. These aspects have led to great variability in lesion assessment amongst clinicians. In this paper, we describe a segmentation algorithm for PET lesions which yields objective segmentations without operator variability. The technique is based on the mean shift algorithm, applied in a spherical coordinate frame to yield a directional assessment of foreground and background and a varying background model. We analyze the algorithm using clinically relevant hybrid digital phantoms and illustrate its effectiveness relative to other techniques.


Segmentation Algorithm Radial Line Shift Algorithm Piecewise Constant Approximation Lesion Boundary 
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.


  1. 1.
    Otsu, N.: A threshold selection method from gray level histograms. T-SMC 1, 62–66 (1979)Google Scholar
  2. 2.
    Dempster, A., Laird, N., Rubin, D.: Maximum likelihood from incomplete data via the EM algorithm. Journal of Royal Statistical Society, Series B 39, 1–38 (1977)MATHMathSciNetGoogle Scholar
  3. 3.
    Ambroise, C., Dang, M., Govaert, G.: Clustering of spatial data by the EM algorithm. In: Geostatistics for Environmental Applications. Quantitative Geology and Geostatistics, vol. 9, pp. 493–504. Kluwer Academic Publishers, Dordrecht (1997)Google Scholar
  4. 4.
    Tolias, Y., Panas, S.: Image segmentation by a fuzzy clustering algorithm using adaptive spatially constrained membership functions. T-SMC 28(3), 359–369 (1998)Google Scholar
  5. 5.
    Manjeshwar, R.M., Akhurst, T.J., Beattie, B.B., Cline, H.E., Gonzalez Trotter, D.E., Humm, J.L.: Realistic hybrid digital phantoms for evaluation of image analysis tools in positron emission tomography. In: SNM (2002)Google Scholar
  6. 6.
    Fukunaga, K., Hostetler, L.D.: The estimation of the gradient of a density function, with applications in pattern recognition. T-IT 21(1), 32–40 (1975)MATHMathSciNetGoogle Scholar
  7. 7.
    Cheng, Y.: Mean shift, mode seeking and clustering. T-PAMI 17(8), 790–799 (1995)Google Scholar
  8. 8.
    Comaniciu, D., Meer, P.: Mean shift: A robust approach toward featue space analysis. T-PAMI 24(5), 603–619 (2002)Google Scholar
  9. 9.
    Comaniciu, D., Ramesh, V., Meer, P.: Real-time tracking of non-rigid objects using mean shift. In: CVPR 2000, vol. 2, pp. 142–149 (2000)Google Scholar
  10. 10.
    Tek, H., Comaniciu, D., Williams, J.: Vessel detection by mean-shift based ray propagation. In: MMBIA, pp. 228–235 (2001)Google Scholar
  11. 11.
    Dice, L.R.: Measures of the amount of ecologic association between species. Ecology 26, 297–302 (1945)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Thomas B. Sebastian
    • 1
  • Ravindra M. Manjeshwar
    • 1
  • Timothy J. Akhurst
    • 2
  • James V. Miller
    • 1
  1. 1.GE ResearchNiskayuna
  2. 2.Department of Nuclear MedicineMSK Cancer CenterNew York

Personalised recommendations