Neural Network Analysis of Dynamic Contrast-Enhanced MRI Mammography

  • Axel Wismüller
  • Oliver Lange
  • Dominik R. Dersch
  • Klaus Hahn
  • Gerda L. Leinsinger
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2130)

Abstract

We present a neural network clustering approach to the analysis of dynamic contrast-enhanced magnetic resonance imaging (MRI) mammography time-series. In contrast to conventional extraction of a few voxel-based perfusion parameters, neural network clustering does not discard information contained in the complete signal dynamics time-series data. We performed exploratory data analysis in patients with breast lesions classified as indeterminate from clinical findings and conventional X-ray mammography. Minimal free energy vector quantization provided a self-organized segmentation of voxels w.r.t. fine-grained differences of signal amplitude and dynamics, thus identifying the lesions from surrounding tissue and enabling a subclassification within the lesions with regard to regions characterized by different MRI signal time-courses. We conclude that neural network clustering can provide a useful extension to the conventional visual inspection of interactively defined regions-of-interest. Thus, it can contribute to the diagnosis of indeterminate breast lesions by non-invasive imaging.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    D.R. Dersch. Eigenschaften neuronaler Vektorquantisierer und ihre Anwendung in der Sprachverarbeitung. Verlag Harri Deutsch, Reihe Physik, Bd. 54, Thun, Frankfurt am Main, 1996. ISBN 3-8171-1492-3.Google Scholar
  2. 2.
    D.R. Dersch and P. Tavan. Load balanced vector quantization. In Proceedings of the International Conference on Artificial Neural Networks ICANN, pages 1067–1070. Springer, 1994.Google Scholar
  3. 3.
    C.K. Kuhl, P. Mielcareck, S. Klaschik, C. Leutner, E. Wardelmann, J. Gieseke, and H.H. Schild. Dynamic breast MR imaging: Are signal intensity time course data useful for differential diagnosis of enhancing lesions? Radiology, 211:101–110, 1999.Google Scholar
  4. 4.
    G.L. Leinsinger, M. Scherr, A. Wismüller, O. Lange, D.R. Dersch, and K. Hahn. Unsupervised cluster analysis of dynamic contrast-enhanced MR mammography by neural networks-characterization of benign and malignant breast lesions. In preparation, 2000.Google Scholar
  5. 5.
    K. Rose, E. Gurewitz, and G.C. Fox. Vector quantization by deterministic annealing. IEEE Transactions on Information Theory, 38(4): 1249–1257, 1992.MATHCrossRefGoogle Scholar
  6. 6.
    A. Wismüller, D.R. Dersch, B. Lipinski, K. Hahn, and D. Auer. Hierarchical clustering of fMRI time-series by deterministic annealing. Neuroimage, 7(4):593, 1998.Google Scholar
  7. 7.
    A. Wismüller, G.L. Leinsinger, D.R. Dersch, O. Lange, M. Scherr, and K. Hahn. Neural network image processing in contrast-enhanced MRI mammography-unsupervised cluster analysis of signal dynamics time-series by deterministic annealing. Radiology Suppl., 217(P):208–209, 2000.Google Scholar
  8. 8.
    A. Wismüller, F. Vietze, and D.R. Dersch. Segmentation with neural networks. In Isaac Bankman, Raj Rangayyan, Alan Evans, Roger Woods, Elliot Fishman, and Henry Huang, editors, Handbook of Medical Imaging, Johns Hopkins University, Baltimore, 2000. Academic Press. ISBN 0120777908.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Axel Wismüller
    • 1
  • Oliver Lange
    • 1
  • Dominik R. Dersch
    • 2
  • Klaus Hahn
    • 1
  • Gerda L. Leinsinger
    • 1
  1. 1.Institut für Radiologische DiagnostikLudwig-Maximilians-Universität MünchenMünchenGermany
  2. 2.Hypovereinsbank AGMünchenGermany

Personalised recommendations