Automatic Contrast Phase Estimation in CT Volumes

  • Michal Sofka
  • Dijia Wu
  • Michael Sühling
  • David Liu
  • Christian Tietjen
  • Grzegorz Soza
  • S. Kevin Zhou
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6893)

Abstract

We propose an automatic algorithm for phase labeling that relies on the intensity changes in anatomical regions due to the contrast agent propagation. The regions (specified by aorta, vena cava, liver, and kidneys) are first detected by a robust learning-based discriminative algorithm. The intensities inside each region are then used in multi-class LogitBoost classifiers to independently estimate the contrast phase. Each classifier forms a node in a decision tree which is used to obtain the final phase label. Combining independent classification from multiple regions in a tree has the advantage when one of the region detectors fail or when the phase training example database is imbalanced. We show on a dataset of 1016 volumes that the system correctly classifies native phase in 96.2% of the cases, hepatic dominant phase (92.2%), hepatic venous phase (96.7%), and equilibrium phase (86.4%) in 7 seconds on average.

Keywords

Contrast Phase Anatomical Region Focal Liver Lesion Compute Tomography Volume Native Phase 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Clunie, D.: Impact of new DICOM objects on handling large data sets. In: University of Rochester PACS 2006 Conference, San Antonio, Texas (March 2006)Google Scholar
  2. 2.
    Deng, X., Du, G.: Editorial: Liver tumor segmentation. In: 3D Segmentation in the Clinic: A Grand Challenge II, MICCAI Workshop, New York, NY (2008)Google Scholar
  3. 3.
    Friedman, J., Hastie, T., Tibshirani, R.: Additive logistic regression: A statistical view of boosting. Annals of Statistics 28(4), 337–407 (2000)MathSciNetCrossRefMATHGoogle Scholar
  4. 4.
    Graser, A., Staehler, M., Reiser, M.F., Mueller-Lisse, U.G.: Low-dose multiphasic protocols offer the best option for cancer detection, classification, and staging. Diagnostic Imaging (April 14, 2006)Google Scholar
  5. 5.
    Laghi, A., Sansoni, I., Celestre, M., Paolantonio, P., Passariello, R.: Computed tomography. In: Lencioni, R., Cioni, D., Bartolozzi, C. (eds.) Focal Liver Lesions: Detection, Characterization, Ablation. Springer, Heidelberg (2005)Google Scholar
  6. 6.
    Meyer, C., Park, H., Balter, J., Bland, P.: Method for quantifying volumetric lesion change in interval liver CT examinations. IEEE TMI 22(6), 776–781 (2003)Google Scholar
  7. 7.
    Nocedal, J., Wright, S.J.: Numerical Optimization. Springer, Heidelberg (2000)MATHGoogle Scholar
  8. 8.
    Sofka, M., Zhang, J., Zhou, S., Comaniciu, D.: Multiple object detection by sequential Monte Carlo and hierarchical detection network. In: CVPR (2010)Google Scholar
  9. 9.
    Tu, Z.: Probabilistic boosting-tree: Learning discriminative models for classification, recognition, and clustering. In: CVPR, vol. 2, pp. 1589–1596 (2005)Google Scholar
  10. 10.
    Valette, P.J., Pilleul, F., Crombé-Ternamian, A.: Imaging benign and metastatic liver tumors with MDCT. In: Marchal, G., Vogl, T., Heiken, J.P. (eds.) Multidetector-Row Computed Tomography: Scanning and Contrast Protocols. Springer, Heidelberg (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Michal Sofka
    • 1
  • Dijia Wu
    • 1
  • Michael Sühling
    • 1
  • David Liu
    • 1
  • Christian Tietjen
    • 2
  • Grzegorz Soza
    • 2
  • S. Kevin Zhou
    • 1
  1. 1.Image Analytics and InformaticsSiemens Corporate ResearchPrincetonUSA
  2. 2.Computed TomographySiemens HealthcareForchheimGermany

Personalised recommendations