European Journal of Nuclear Medicine

, Volume 21, Issue 12, pp 1285–1292 | Cite as

Principal component analysis of dynamic positron emission tomography images

  • F. Pedersen
  • M. Bergströme
  • E. Bengtsson
  • B. Långström
Original Article


Multivariate image analysis can be used to analyse multivariate medical images. The purpose could be to visualize or classify structures in the image. One common multivariate image analysis technique which can be used for visualization purposes is principal component analysis (PCA). The present work concerns visualization of organs and structures with different kinetics in a dynamic sequence utilizing PCA. When applying PCA on positron emission tomography (PET) images, the result is initially not satisfactory. It is illustrated that one major explanation for the behaviour of PCA when applied to PET images is that it is a data-driven technique which cannot separate signals from high noise levels. With a better understanding of the PCA, gained with a strategy of examining the image data set, the transformations, and the results using visualization tools, a surprisingly easily understood methodology can be derived. The proposed methodology can enhance clinically interesting information in a dynamic PET imaging sequence in the first few principal component images and thus should be able to aid in the identification of structures for further analysis.

Key words

PET imaging Multivariate image analysis Principal component analysis Visualization of multidimensional data 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Duda R, Hart P.Pattern classification an scene analysis. New York: Wiley, 1973.Google Scholar
  2. 2.
    Draper N, Smith H.Applied regression analysis, 2nd edn. New York: Wiley, 1981.Google Scholar
  3. 3.
    Harmon H.Modern factor analysis, 2nd edn. Chicago: The University of Chicago Press, 1967.Google Scholar
  4. 4.
    Ready P, Wintz P. Information extraction, SNR improvement, and data compression in multispectral imagery,IEEE Trans Commun 1973; 21: 1123–1131.CrossRefGoogle Scholar
  5. 5.
    Ortendahl D. The application of principal component analysis to mulitvariate MRI data.Proceedings of the 8th Annual Conference of the Engineering in Medicine and Biology Society Fort Worth, Texas 1986; 1065–1068.Google Scholar
  6. 6.
    Higer H, Bielke G, eds.Tissue characterization in MR imaging. Clinical and technical approaches. Berlin Heidelberg New York: Springer, 1990.Google Scholar
  7. 7.
    Barber D. The use of principal components in the quantitative analysis of gamma camera dynamic studies.Phys Med Biol 1980; 25: 283–292.PubMedCrossRefGoogle Scholar
  8. 8.
    Frouin F, Bazin J-P, Di Paola R. Image sequence processing using factor analysis and compartmental modelling.SPIE Vol 1137 Science and Engineering of Medical Images 1989; 37–44.Google Scholar
  9. 9.
    Jackson J.A user's guide to principal components. New York: Wiley, 1991.Google Scholar
  10. 10.
    Johnson R, Wichern D.Applied multivariate statistical analysis, 2nd edn. Englewoods Cliffs, N.J.: Prentice-Hall, 1988.Google Scholar
  11. 11.
    Jolliffe I.Principal component analysis. New York Berlin Heidelberg: Springer, 1986.Google Scholar
  12. 12.
    Barber D, Martel A. Factor analysis revisited.Eur J Nucl Med 1992; 19: 467–468.PubMedCrossRefGoogle Scholar
  13. 13.
    Pedersen F, Bengtsson E, Jonsson D. A numerically derived method for preprocessing of noisy data before applying principal component analysis,Proceedings of the 8th Scandinavian Conference on Image Analysis. Norwegian Society for Image Processing and Pattern Recognition, Tromsø 1993; Vol II: 981–988.Google Scholar
  14. 14.
    Pedersen F, Bergstrom M, Bengtsson E, Maripuu E. Principal component analysis of dynamic PET and gamma camera images: A methodology to visualize the signals in the presence of large noise.1993 IEEE Conference record Nuclear Science Symposium and Medical Imaging Conference. San Francisco, Calif. 1993; Vol 3: 1734–1738.CrossRefGoogle Scholar
  15. 15.
    Green A, Berman M, Switzer P, Craig M. A transformation for ordering mutispectral data in terms of image quality with implications for noise removal.IEEE Trans Geosci Remote Sensing 1988; 26: 65–74.CrossRefGoogle Scholar
  16. 16.
    Bengtsson E, Nordin B, Pedersen F. MUSE — a new tool for interactive image analysis and segmentation based on multivariate statistics.Comput Methods Programs Biomed 1994; 42: 181–200.PubMedCrossRefGoogle Scholar
  17. 17.
    Pedersen E.Interactive explorative analysis of multivariate images using principal components. PhD thesis, Centre for Image Analysis, Uppsala University, Sweden, 1994.Google Scholar
  18. 18.
    Esbensen K, Geladi P. Strategy of multivariate image analysis (MIA).Chemometrics and Intelligent Laboratory Systems 1989; 7: 67–86.CrossRefGoogle Scholar
  19. 19.
    Eriksson B, Bergstrom M, Lilja A, Ahlström H, Långström B, Öberg K. Positron emission tomography (PET) in neuroendocrine gastrointestinal tumors.Acta Oncol 1993; 32: 189–196.PubMedGoogle Scholar
  20. 20.
    Gonzalez R, Wintz P.Digital image processing, 2nd edn. Reading, Mass.: Addison-Wesley, 1987.Google Scholar
  21. 21.
    Geladi P, Isaksson H, Lindqvist L, Wold S, Esbensen K. Principal component analysis of multivariate images.Chemometrics and Intelligent Laboratory Systems 1989; 5: 209–220.CrossRefGoogle Scholar
  22. 22.
    Press W, Flannery B, Teukolsky S, Vetterling W.Numerical recipes in C. Cambridge: Cambridge University Press, 1988.Google Scholar

Copyright information

© Springer-Verlag 1994

Authors and Affiliations

  • F. Pedersen
    • 1
  • M. Bergströme
    • 2
  • E. Bengtsson
    • 1
  • B. Långström
    • 2
  1. 1.Centre for Image AnalysisUppsala UniversityUppsalaSweden
  2. 2.University HospitalPET Center, Uppsala UniversityUppsalaSweden

Personalised recommendations