Advertisement

Epidural Masses Detection on Computed Tomography Using Spatially-Constrained Gaussian Mixture Models

  • Sanket Pattanaik
  • Jiamin LiuEmail author
  • Jianhua Yao
  • Weidong Zhang
  • Evrim Turkbey
  • Xiao Zhang
  • Ronald Summers
Conference paper
Part of the Lecture Notes in Computational Vision and Biomechanics book series (LNCVB, volume 17)

Abstract

The widespread use of CT imaging and the critical importance of early detection of epidural masses of the spinal canal generate a scenario ideal for the implementation of a computer-aided detection (CAD) system. Epidural masses can lead to paralysis, incontinence and loss of neurological function if not promptly detected. We present, to our knowledge, the first CAD system to detect epidural masses on CT. In this paper, global intensity and local spatial features are modeled as spatially constrained Gaussian Mixture Model (CGMM) for epidural mass detection. The Cross-validation on 23 patients with epidural masses on body CT showed that the CGMM yielded a marked improvement of performance (69 % at 8.6 false positives per patient) over an intensity based K-means method (46 % at 7.9 false-positives per patient).

Keywords

Spinal Canal Gaussian Mixture Model Local Binary Pattern Gaussian Component Tissue Class 
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.
    Schiff, D., O’Neill, B.P., et al.: Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology 49(2), 452–456 (1997)CrossRefGoogle Scholar
  2. 2.
    Hricak, H., Akin, O., Bradbury, M.S.: Functional and metabolic imaging. In: DeVita, V.T. (ed.) Cancer: Principles & Practice of Oncology, pp. 589–616. Williams & Wilkins, Baltimore (2005)Google Scholar
  3. 3.
    Fine, H., Barker, F.G., Markert, J.M., Loeffler, J.S.: Neoplasms of the central nervous system. Cancer: Principles & Practice of Oncology, pp. 1834–1887. Williams & Wilkins, Baltimore (2005)Google Scholar
  4. 4.
    Freifeld, O., Greenspan, H.: Multiple sclerosis lesion detection using constrained GMM and curve evolution. Int. J. Biomed. Imaging 2009, 14 (2009)CrossRefGoogle Scholar
  5. 5.
    Jianhua, Y., O’Connor, S. D., et al.: Automated spinal column extraction and partitioning. In: 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, pp 390–393 (2006)Google Scholar
  6. 6.
    Haralick, R.M., Shanmugam, K., Dinstein, I.: Textural features for image classification. IEEE Trans. Syst. Man Cybern. 3, 610–621 (1973)Google Scholar
  7. 7.
    Yao, J., Summers, R.M., et al.: Optimizing the support vector machines (SVM) committee configuration in a colonic polyp CAD system. Proc. SPIE 5746, 384–392 (2005)CrossRefGoogle Scholar
  8. 8.
    Carlson, J., Heckerman, D., Shani, G.: False discovery rate for 2x2 contingency tables. Microsoft Research technical report MSR-TR-2009-53 (2009)Google Scholar
  9. 9.
    Wang, Q.: HMRF-EM-image: implementation of the Hidden Markov random field model and its expectation-maximization algorithm. CoRR e-prints (1207.3510)Google Scholar
  10. 10.
    Dorwart, R.H., DeGroot, J., Sauerland, E.T., et al.: CT of the lumbar spine: normal variants and pitfalls. Radiographics 2(4), 459–499 (1982)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Sanket Pattanaik
    • 1
  • Jiamin Liu
    • 1
    Email author
  • Jianhua Yao
    • 1
  • Weidong Zhang
    • 1
  • Evrim Turkbey
    • 1
  • Xiao Zhang
    • 1
  • Ronald Summers
    • 1
  1. 1.Radiology and Imaging Sciences Department, Clinical CenterThe National Institutes of HealthBethesdaUSA

Personalised recommendations