Multi-stage Learning for Robust Lung Segmentation in Challenging CT Volumes

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6893)


Simple algorithms for segmenting healthy lung parenchyma in CT are unable to deal with high density tissue common in pulmonary diseases. To overcome this problem, we propose a multi-stage learning-based approach that combines anatomical information to predict an initialization of a statistical shape model of the lungs. The initialization first detects the carina of the trachea, and uses this to detect a set of automatically selected stable landmarks on regions near the lung (e.g., ribs, spine). These landmarks are used to align the shape model, which is then refined through boundary detection to obtain fine-grained segmentation. Robustness is obtained through hierarchical use of discriminative classifiers that are trained on a range of manually annotated data of diseased and healthy lungs. We demonstrate fast detection (35s per volume on average) and segmentation of 2 mm accuracy on challenging data.


Interstitial Lung Disease Shape Model Statistical Shape Model Lung Surface Mesh Vertex 
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.


  1. 1.
    De Nunzio, G., Tommasi, E., Agrusti, A., Cataldo, R., De Mitri, I., Favetta, M., Maglio, S., Massafra, A., Quarta, M., Torsello, M., et al.: Automatic lung segmentation in CT images with accurate handling of the hilar region. Journal of Digital Imaging, 1–17 (2009)Google Scholar
  2. 2.
    Hu, S., Hoffman, E., Reinhardt, J.: Automatic lung segmentation for accurate quantitation of volumetric X-ray CT images. IEEE Trans. Med. Imag. 20(6), 490–498 (2002)CrossRefGoogle Scholar
  3. 3.
    Kockelkorn, T., van Rikxoort, E., Grutters, J., van Ginneken, B.: Interactive lung segmentation in CT scans with severe abnormalities. In: IEEE Int. Symp. on Biomed. Imag., pp. 564–567 (2010)Google Scholar
  4. 4.
    Korfiatis, P., Kalogeropoulou, C., Karahaliou, A., Kazantzi, A., Skiadopoulos, S., Costaridou, L.: Texture classification-based segmentation of lung affected by interstitial pneumonia in high-resolution CT. Medical Physics 35, 5290 (2008)CrossRefGoogle Scholar
  5. 5.
    Ling, H., Zhou, S.K., Zheng, Y., Georgescu, B., Suehling, M., Comaniciu, D.: Hierarchical, learning-based automatic liver segmentation. In: IEEE Conf. on Comp. Vis. and Patt. Recog., Los Alamitos, CA, USA (2008)Google Scholar
  6. 6.
    Prasad, M., Brown, M., Ahmad, S., Abtin, F., Allen, J., da Costa, I., Kim, H., McNitt-Gray, M., Goldin, J.: Automatic segmentation of lung parenchyma in the presence of diseases based on curvature of ribs. Acad. Radiol. 15(9), 1173–1180 (2008)CrossRefGoogle Scholar
  7. 7.
    Sluimer, I., Prokop, M., van Ginneken, B.: Toward automated segmentation of the pathological lung in CT. IEEE Trans. Med. Imag. 24(8), 1025–1038 (2005)CrossRefGoogle Scholar
  8. 8.
    Sofka, M., Zhang, J., Zhou, S., Comaniciu, D.: Multiple object detection by sequential Monte Carlo and hierarchical detection network. In: IEEE Conf. on Comp. Vis. and Patt. Recog. (June 13-18, 2010)Google Scholar
  9. 9.
    Sun, S., McLennan, G., Hoffman, E.A., Beichel, R.: Model-based segmentation of pathological lungs in volumetric CT data. In: The Third International Workshop on Pulmonary Image Analysis (2010)Google Scholar
  10. 10.
    Tu, Z.: Probabilistic boosting-tree: learning discriminative models for classification, recognition, and clustering. In: IEEE Int. Conf. on Comp. Vis. and Patt. Recog., vol. 2, pp. 1589–1596 (2005)Google Scholar
  11. 11.
    Vidal, C., Hewitt, J., Davis, S., Younes, L., Jain, S., Jedynak, B.: Template registration with missing parts: Application to the segmentation of M. tuberculosis infected lungs. In: IEEE Int. Symp. on Biomed. Imag., pp. 718–721 (2009)Google Scholar
  12. 12.
    Wang, J., Li, F., Li, Q.: Automated segmentation of lungs with severe interstitial lung disease in CT. Medical Physics 36, 4592 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  1. 1.Image Analytics and InformaticsSiemens Corporate ResearchPrincetonUSA
  2. 2.Molecular ImagingSiemens HealthcareOxfordUK

Personalised recommendations