Skip to main content
SpringerLink
Log in

HOSVD-Based 3D Active Appearance Model: Segmentation of Lung Fields in CT Images

  • Systems-Level Quality Improvement
  • Published:
Journal of Medical Systems Aims and scope Submit manuscript

Abstract

An Active Appearance Model (AAM) is a computer vision model which can be used to effectively segment lung fields in CT images. However, the fitting result is often inadequate when the lungs are affected by high-density pathologies. To overcome this problem, we propose a Higher-order Singular Value Decomposition (HOSVD)-based Three-dimensional (3D) AAM. An evaluation was performed on 310 diseased lungs form the Lung Image Database Consortium Image Collection. Other contemporary AAMs operate directly on patterns represented by vectors, i.e., before applying the AAM to a 3D lung volume,it has to be vectorized first into a vector pattern by some technique like concatenation. However, some implicit structural or local contextual information may be lost in this transformation. According to the nature of the 3D lung volume, HOSVD is introduced to represent and process the lung in tensor space. Our method can not only directly operate on the original 3D tensor patterns, but also efficiently reduce the computer memory usage. The evaluation resulted in an average Dice coefficient of 97.0 % ± 0.59 %, a mean absolute surface distance error of 1.0403 ± 0.5716 mm, a mean border positioning errors of 0.9187 ± 0.5381 pixel, and a Hausdorff Distance of 20.4064 ± 4.3855, respectively. Experimental results showed that our methods delivered significant and better segmentation results, compared with the three other model-based lung segmentation approaches, namely 3D Snake, 3D ASM and 3D AAM.

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

Access this article

Log in via an institution

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

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

References

  1. Cao, P., Yang, J. Z., Li, W., et al., Ensemble-based hybrid probabilistic sampling for imbalanced data learning in lung nodule CAD. Comput Med Imaging Graph 38:137–150, 2014.

    Article  PubMed  Google Scholar 

  2. Tasci, E., and Ugur, A., Shape and texture based novel features for automated juxtapleural nodule detection in lung CTs. J Med Syst 39(46):1–13, 2015.

    Google Scholar 

  3. Singh, S. P., and Urooi, S., An improved CAD system for breast cancer diagnosis based on generalized pseudo-Zernike moment and Ada-DEWNN classifier. J Med Syst 40(4):1–13, 2016.

    Article  CAS  Google Scholar 

  4. Cetin, M., and Iskurt, A., An automatic 3-D reconstruction of coronary arteries by stereopsis. J Med Syst 40(4):1–11, 2016.

    Article  Google Scholar 

  5. Chen, X. J., Udupa, J. K., Bagci, U., et al., Medical image segmentation by combining graph cuts and oriented active appearance models. IEEE Trans Image Process 21(4):2035–2046, 2012.

    Article  PubMed  Google Scholar 

  6. Dawoud, A., Lung segmentation in chest radiographs by fusing shape information in iterative thresholding. IET Comput Vis 5(3):185–190, 2011.

    Article  Google Scholar 

  7. Noor, N. M., Than, J. C. M., Rijal, O. M., et al., Automatic lung segmentation using control feedback system: Morphology and texture paradigm. J Med Syst 39(3):1–18, 2015.

    Article  Google Scholar 

  8. Bae, K. T., Kim, J. S., Na, Y. H., et al., Pulmonary nodules: Automated detection on CT images with morphologic matching algorithm-preliminary results [J]. Radiology 236:286–294, 2005.

    Article  PubMed  Google Scholar 

  9. Li, B., and Acton, S. T., Automatic active model initialization via Poisson inverse gradient. IEEE Trans Image Process 17(8):1406–1419, 2008.

    Article  PubMed  Google Scholar 

  10. Norliza, M. N., Joel, C. M. T., and Omar, M. R., Automatic lung segmentation using control feedback system: Morphology and texture paradigm. J Med Syst 39(22):1–18, 2015.

    Google Scholar 

  11. Pu, J., Roos, C. A. Y., Napel, S., et al., Adaptive border marching algorithm : Automatic lung segmentation on chest CT images. Comput Med Imaging Graph 32(6):452–462, 2008.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Wang, J., Li, Q., Li, F., et al., Automated segmentation of lungs with severe interstitial lung disease in CT. Med Physics 36(1):4592–4599, 2009.

    Article  Google Scholar 

  13. Rikxoort, E. M. V., Hoop, B. D., Viergever, M. A., et al., Automatic lung segmentation from thoracic computed tomography scans using a hybrid approach with error detection. Med Phys 36:2934–2947, 2009.

    Article  PubMed  Google Scholar 

  14. Rikxoort, E. M. V., and Ginneken, B. V., Automated segmentation of pulmonary structures in thoracic computed tomography scans: A review. Phys Med Biol 58(17):R187, 2013.

    Article  PubMed  Google Scholar 

  15. Xu, T., Mandal, M., Long, R., et al., An edge-region force guided active shape approach for automatic lung field detection in chest radiographs. Comput Med Imaging Graph 36(6):452–463, 2012.

    Article  PubMed  Google Scholar 

  16. Liu, J., and Udupa, J., Oriented active shape models. IEEE Trans Med Imaging 28(4):571–584, 2009.

    Article  CAS  PubMed  Google Scholar 

  17. Sun, S. H., Bauer, C., and Beichel, R., Automated 3-D segmentation of lungs with lung cancer in CT data using a novel robust active shape model approach. IEEE Trans Med Imaging 31(2):449–460, 2012.

    Article  CAS  PubMed  Google Scholar 

  18. Catalina, T. G., Federico, M. S., Constantine, B., et al., Automatic training and reliability estimation for 3D ASM applied to cardiac MRI segmentation. Phys Med Biol 57:4155–4174, 2012.

    Article  Google Scholar 

  19. Steven, C. M., Johan, G. B., Boudewijn, P. F. L., et al., 3-D active appearance models: Segmentation of cardiac MR and ultrasound images. IEEE Trans Med Imaging 21(9):1167–1178, 2002.

    Article  Google Scholar 

  20. Larsen, R., Stegmann, M., Darkner, S., et al., Texture enhanced appearance models. Comput Vis Image Underst 106(1):20–30, 2007.

    Article  Google Scholar 

  21. Baka, N., Milles, J., Hendriks, E., et al., Segmentation of myocardial perfusion MR sequences with multi-band active appearance models driven by spatial and temporal features. Proc SPIE Med Imaging 6914:1–10, 2008.

    Google Scholar 

  22. Toth, R., and Madabhushi, A., Multifeature landmark-free active appearance models: Application to prostate MRI segmentation. IEEE Trans Med Imaging 31(8):1638–1650, 2012.

    Article  PubMed  Google Scholar 

  23. Toth, R., Ribault, J., Gentile, J., et al., Simultaneous segmentation of prostatic zones using active appearance models with multiple coupled levelsets. Comput Vis Image Underst 117:1051–1060, 2013.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Cootes, T. F., Taylor, C. J., Cooper, D. H., et al., Active shape models-their training and application. Comput Vis Image Underst 61(1):38–59, 1995.

    Article  Google Scholar 

  25. Cootes, T. F., Edwards, G. J., and Taylor, C. J., Active appearance models. IEEE Trans Pattern Anal Mach Intell 23(6):681–685, 2001.

    Article  Google Scholar 

  26. Vasilescu M. A. O., Terzopoulos D. Multilinear Subspace Analysis of Image Ensembles, CVPR. 2003: 93–99

  27. Lu, H. P., Konstantinos, N. P., and Anastasios, N. V., MPCA: Multilinear principal component analysis of tensor objects. IEEE Trans. Neural Netw 19(1):18–39, 2008.

    Article  PubMed  Google Scholar 

  28. Nie, F., Xiang, S., Song, Y., et al., Extracting the optimal dimensionality for local tensor discriminant analysis. Pattern Recogn 42(1):105–114, 2009.

    Article  Google Scholar 

  29. Tao, D., Li, X., Wu, X., et al., Supervised tensor learning [J]. Knowl Inf Syst 13(1):1–42, 2007.

    Article  Google Scholar 

  30. Tao, D., Li, X., Wu, X., et al., Tensor rank One discriminant analysis-a convergent method for discriminative multilinear subspace selection. Neurocomputing 7110:1866–1882, 2008.

    Article  Google Scholar 

  31. Wang, Q. Z., Zhu, W. C., and Wang, B., Three-dimensional SVM with latent variable: Application for detection of lung lesions in CT images. J Med Syst 39(1):1–7, 2015.

    Google Scholar 

  32. Hou, C. P., Nie, F. P., Zhang, C. S., et al., Multiple rank multi-linear SVM for matrix date classification. Pattern Recogn 47:454–469, 2014.

    Article  Google Scholar 

  33. Feng Z. H., Kittler J., Christmas W., et al. Automatic Face Annotation by Multilinear AAM with Missing Values. 21st International Conference on Pattern Recognition. 2012:11–15

  34. Lieven, D. L., et al., A multilinear singular value decomposition. Siam J Matrix Anal Appl 21(4):1253–1278, 2000.

    Article  Google Scholar 

  35. Vannieuwenhoven, N., et al., A New truncation strategy for the higher-order singular value decomposition. Siam J Sci Comput 34(2):1027–1052, 2012.

    Article  Google Scholar 

  36. Heimann, T., Ginneken, B. V., Styner, M. A., et al., Comparison and evaluation of methods for liver segmentation from CT datasets. IEEE Trans Med Imaging 28(8):1251–1265, 2009.

    Article  PubMed  Google Scholar 

Download references

Acknowledgment

This work was supported by National Natural Science Foundation of China (61301257). The experiment data are from LIDC and Jilin Tumor Hospital.

Author information

Authors and Affiliations

  1. School of Information Engineering, Northeast Dianli University, Jilin, 132012, China

    Qingzhu Wang, Wanjun Kang & Haihui Hu

  2. Jilin Tumor Hospital, Changchun, China

    Bin Wang

Authors
  1. Qingzhu Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wanjun Kang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Haihui Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Bin Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Qingzhu Wang.

Additional information

This article is part of the Topical Collection on Systems-Level Quality Improvement

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, Q., Kang, W., Hu, H. et al. HOSVD-Based 3D Active Appearance Model: Segmentation of Lung Fields in CT Images. J Med Syst 40, 176 (2016). https://doi.org/10.1007/s10916-016-0535-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10916-016-0535-0

Keywords

Search

Navigation