Advertisement

Thin-Plate splines and the atlas problem for biomedical images

  • Fred L. Bookstein
6. Anatomical Models And Variability
Part of the Lecture Notes in Computer Science book series (LNCS, volume 511)

Abstract

The thin-plate spline, a general-purpose interpolant for labelled point data, ties the geometry of image deformation to the classic biometric algebra of quadratic forms. The technique thus helps not only in the production of biomedical atlases—averaged or normative images of particular structures or configurations of parts—but also in the understanding of specimen-by-specimen variability around those atlases. This report summarizes the statistics of thirteen landmark points in midsagittal MRI images of nine normal adult human brains, produces and evaluates averaged images with the aid of those statistics, and pursues some implications for stereotaxy, for region-tracing, and for the understanding of averaged image structures.

Keywords

Image deformation image warping image averaging landmark data MRI morphometrics neuroanatomy anatomical norms brain shape 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bookstein FL (1986a). Size and shape spaces for landmark data in two dimensions. Statistical Science 1:181–242.Google Scholar
  2. Bookstein FL (1986b). From medical imaging to the biometrics of form. In: Information Processing in Medical Imaging. Bacharach SL (ed), Martinus Nijhoff, Dordrecht, Netherlands, pp. 1–18.Google Scholar
  3. Bookstein FL (1987). Morphometrics for functional imaging studies. In Mazziotta JC and Koslow SH, Assessment of Goals and Obstacles in Data Acquisition and Analysis from Emission Tomography: Report of a Series of International Workshops. Journal of Cerebral Blood Flow and Metabolism 7:S23–S27.Google Scholar
  4. Bookstein FL (1988). Toward a notion of feature extraction In: Information Processing in Medical Imaging. de Graaf CN and Viergever MA (eds), Plenum Press, New York, pp. 23–43.Google Scholar
  5. Bookstein FL (1989). Principal warps: Thin-plate splines and the decomposition of deformations. I.E.E.E. Transactions on Pattern Analysis and Machine Intelligence PAMI-11:567–585.Google Scholar
  6. Bookstein FL(1990a). Morphometrics. In Toga, AW (ed), Raven Press, New York, pp. 167–188.Google Scholar
  7. Bookstein FL (1990b). Distortion correction. In: Three-Dimensional Neuroimaging. Toga, AW (ed), Raven Press, New York, pp. 235–249.Google Scholar
  8. Bookstein FL (1991a). Four metrics for image variation. In: Information Processing in Medical Imaging. Ortendahl DA and Llacer J (eds), Wiley-Liss, New York, pp. 227–240.Google Scholar
  9. Bookstein FL (1991b). Morphometric Tools for Landmark Data. Cambridge University Press, New York, to appear.Google Scholar
  10. Bookstein FL and Jaynes W (1990). Thin-plate splines and the analysis of biological shape. Video tape, 21 minutes.Google Scholar
  11. Bookstein FL, Mazziotta JC and Valentino D (1991). Morphometrics for brain atlases: landmarks, deformations, and picture averaging. NeuroImage, submitted.Google Scholar
  12. Fox PT and Kall B (1987). Stereotaxy as a means of anatomical localization in physiological brain images: Proposals for further validation. In Mazziotta JC and Koslow SH, Assessment of Goals and Obstacles in Data Acquisition and Analysis from Emission Tomography: Report of a Series of International Workshops. Journal of Cerebral Blood Flow and Metabolism 7:S18–S20.Google Scholar
  13. Rohlf FJ and Bookstein FL (1990). Proceedings of the Michigan Morphometrics Workshop. University of Michigan Museums, Ann Arbor, MI.Google Scholar
  14. Terzopoulos D (1983). Multilevel computational processes for visual surface reconstruction. Computer Vision, Graphics, and Image Processing 24:52–96.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Fred L. Bookstein
    • 1
  1. 1.Center for Human GrowthUniversity of MichiganAnn ArborUSA

Personalised recommendations