Skip to main content
Log in

Enhancement of abnormality in gamma images using Kohonen filters

  • Published:
European Journal of Nuclear Medicine Aims and scope Submit manuscript

Abstract

A new method of processing digital images consists of a numerical filter which suppresses normal features producing an enhancement of the abnormality. The resulting filtered image aids the observer in the recognition of the abnormality. The filter is developed from verified normal images in the computer memory. The enhancement of pathologic accumulations with respect to surrounding activity was improved on the average by a factor of 2.6 for 13 lateral brain images.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  • Barber, D.C., Mallard, J.R.: Data processing of radioisotope images for optimum smoothing. Phys. in Med. Biol. 16 635–644 (1971)

    Google Scholar 

  • Brown, D.W., Kirch, D.L., Ryerson, T.W.: Computer processing of scans using Fourier and other transformations. J. nucl. Med. 12, 287–291 (1971)

    Google Scholar 

  • Delashmit, W.H., Webber, R.F.: Gray level enhancement techniques for increasing the observability of the borders of features of digitized imagery. Photogr. Sci. Eng. 17, 518–523 (1973)

    Google Scholar 

  • Gramenopoulos, N.: Optimization of information display in image recorders. Photogr. Sci. Eng. 18, 658–663 (1974)

    Google Scholar 

  • Herath, K.B., Sharp, P.F.: Effects of “matched filter” smoothing as measured by receiver operating characteristic curve. Phys. in Med. Biol. 21, 442–446 (1976)

    Google Scholar 

  • Kohonen, T.: An adaptive associative memory principle. IEEE Trans. Comp. 23, 444–445 (1974)

    Google Scholar 

  • Kohonen, T.: Associative Memory; A System-Theoretical Approach. Springer-Verlag Series: Communication and Cybernitics 16, (in press)

  • Kohonen, T., Oja, E.: Fast adaptive formation of orthogonalizing filters and associative memory in recurrent networks of neuronlike elements. Biol. Cybern. 21, 85–95 (1976)

    Google Scholar 

  • Kohonen, T., Oja, E., Ruohonen, M.: Adaption of a linear system to finite set of patterns occuring in an arbitrarily varying order. Acta polytech. scand. 25, 7–15 (1974)

    Google Scholar 

  • Kuhl, D.E.: The author's reply. J. nucl. Med. 14, 876 (1973)

    Google Scholar 

  • Kuhl, D.E., Sanders, T.D., Edwards, R.Q.: Failure to improve observer performance with scan smoothing. J. nucl. Med. 13, 752–757 (1972)

    Google Scholar 

  • Metz, C.E., Goodenough, D.J.: On failure to improve observer performance with scan smoothing: A rebuttal. J. nucl. Med. 14, 873–876 (1973)

    Google Scholar 

  • Pratt, W.K.: Generalized Wiener filtering computation techniques. IEEE Trans. Comp. 21, 636–641 (1972)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Riihimäki, E., Häll, LE., Eistola, P. et al. Enhancement of abnormality in gamma images using Kohonen filters. Eur J Nucl Med 1, 259–262 (1976). https://doi.org/10.1007/BF00252174

Download citation

  • Received:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF00252174

Keywords

Navigation