Adaptive template moderated spatially varying statistical classification

  • Simon K. Warfield
  • Michael Kaus
  • Ferenc A. Jolesz
  • Ron Kikinis
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1496)


A novel image segmentation algorithm was developed to allow the automatic segmentation of both normal and abnormal anatomy. The new algorithm is a form of spatially varying classification (SVC), in which an explicit anatomical template is used to moderate the segmentation obtained by k Nearest Neighbour (k-NN) statistical classification. The new algorithm consists of an iterated sequence of spatially varying classification and nonlinear registration, which creates an adaptive, template moderated (ATM), spatially varying classification (SVC).

The ATM SVC algorithm was applied to several segmentation problems, involving different types of imaging and different locations in the body. Segmentation and validation experiments were carried out for problems involving the quantification of normal anatomy (MRI of brains of babies, MRI of knee cartilage of normal volunteers) and pathology of various types (MRI of patients with multiple sclerosis, MRI of patients with brain tumours, MRI of patients with damaged knee cartilage). In each case, the ATM SVC algorithm provided a better segmentation than statistical classification or elastic matching alone.


template moderated segmentation elastic matching nearest neighbour classification knee cartilage neonate brain tumour 


  1. [1]
    S Solloway, CE Hutchinson, JC Waterton, and CJ Taylor, “The use of active shape models for making thickness measurements of articular cartilage from MR images.”, Magn Reson Med, vol. 37, no. 6, pp. 943–952, 1997.CrossRefPubMedGoogle Scholar
  2. [2]
    D. Louis Collins, 3D Model-based segmentation of individual brain structures from magnetic resonance imaging data, PhD thesis, McGill University, 1994.Google Scholar
  3. [3]
    Arun Kumar, Allen R. Tannenbaum, and Gary J. Balas, “Optical Flow: A Curve Evolution Approach”, IEEE Transactions On Image Processing, vol. 5, no. 4, pp. 598–610, April 1996.CrossRefPubMedGoogle Scholar
  4. [4]
    Ron Kikinis, Martha E. Shenton, Dan V. Iosifescu, Robert W. McCarley, Pairash Saiviroonporn, Hiroto H. Hokama, Andre Robatino, David Metcalf, Cynthia G. Wible, Chiara M. Portas, Robert M. Donnino, and Ferenc A. Jolesz, “A Digital Brain Atlas for Surgical Planning, Model Driven Segmentation, and Teaching”, IEEE Transactions on Visualization and Computer Graphics, vol. 2, no. 3, pp. 232–241, September 1996.CrossRefGoogle Scholar
  5. [5]
    G. Gerig, O. Kübler, R. Kikinis, and F. A. Jolesz, “Nonlinear Anisotropic Filtering of MRI Data”, IEEE Transactions On Medical Imaging, vol. 11, no. 2, pp. 221–232, June 1992.CrossRefPubMedGoogle Scholar
  6. [6]
    Simon Warfield, Ferenc Jolesz, and Ron Kikinis, “A High Performance Computing Approach to the Registration of Medical Imaging Data”, Parallel Computing, 1998, To appear in Parallel Computing.Google Scholar
  7. [7]
    Richard O. Duda and Peter E. Hart, Pattern Classification and Scene Analysis, John Wiley & Sons, Inc., 1973.Google Scholar
  8. [8]
    Jerome H. Friedman, Forest Baskett, and Leonard J. Shustek, “An Algorithm for Finding Nearest Neighbors”, IEEE Transactions On Computers, vol. C-24, no. 10, pp. 1000–1006, October 1975.CrossRefGoogle Scholar
  9. [9]
    Simon Warfield, “Fast k-NN Classification for Multichannel Image Data”, Pattern Recognition Letters, vol. 17, no. 7, pp. 713–721, 1996.CrossRefGoogle Scholar
  10. [10]
    Joachim Dengler and Markus Schmidt, “The Dynamic Pyramid — A Model for Motion Analysis with Controlled Continuity”, International Journal of Pattern Recognition and Artificial Intelligence, vol. 2, no. 2, pp. 275–286, 1988.CrossRefGoogle Scholar
  11. [11]
    W. M. Wells, R Kikinis, W. E. L. Grimson, and F. Jolesz, “Actaptive segmentation of MRI data”, IEEE Transactions On Medical Imaging, vol. 15, pp. 429–442, 1996.CrossRefPubMedGoogle Scholar
  12. [12]
    Petra Hüppi, Simon Warfield, Ron Kikinis, Patrick D. Barnes, Gary P. Zientara, Ferenc A. Jolesz, Miles K. Tsuji, and Joseph J. Volpe, “Quantitative Magnetic Resonance Imaging of Brain Development in Premature and Mature Newborns”, Ann Neurol, vol. 43, no. 2, pp. 224–235, February 1998.CrossRefPubMedGoogle Scholar
  13. [13]
    Simon Warfield, Joachim Dengler, Joachim Zaers, Charles R.G. Guttmann, William M. Wells III, Gil J. Ettinger, John Hiller, and Ron Kikinis, “Automatic identification of Grey Matter Structures from MRI to Improve the Segmentation of White Matter Lesions”, Journal of Image Guided Surgery, vol. 1, no. 6, pp. 326–338, 1995.CrossRefPubMedGoogle Scholar
  14. [14]
    Simon Warfield, Carl Winalski, Ferenc Jolesz, and Ron Kikinis, “Automatic Segmentation ofMRI of the Knee”, in ISMRM Sixth Scientific Meeting and Exhibition, April 18–24 1998, p. 563.Google Scholar
  15. [15]
    M. Kamber, D. L. Collins, R. Shinghal, G. S. Francis, and A. C. Evans, “Model-based 3D segmentation of multiple sclerosis lesions in dual-echo MRI data”, in SPIE Vol. 1808, Visualization in Biomedical Computing, 1992, pp. 590–600.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Simon K. Warfield
    • 1
  • Michael Kaus
    • 1
  • Ferenc A. Jolesz
    • 1
  • Ron Kikinis
    • 1
  1. 1.Surgical Planning Laboratory, Harvard Medical School, Department of RadiologyBrigham and Women’s HospitalBoston

Personalised recommendations