Adaptive Voxel, Texture and Temporal Conditional Random Fields for Detection of Gad-Enhancing Multiple Sclerosis Lesions in Brain MRI

  • Zahra Karimaghaloo
  • Hassan Rivaz
  • Douglas L. Arnold
  • D. Louis Collins
  • Tal Arbel
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8151)


The detection of Gad-enhancing lesions in brain MRI of Multiple Sclerosis (MS) patients is of great clinical interest since they are important markers of disease activity. However, many of the enhancing voxels are associated with normal structures (i.e. blood vessels) or noise in the MRI, making the detection of Gad-enhancing lesions a challenging task. Furthermore, these lesions are typically small and in close proximity to vessels. In this paper, we present a probabilistic Adaptive Multi-level Conditional Random Field (AMCRF) framework, capable of leveraging spatial and temporal information, for detection of MS Gad-enhancing lesions. In the first level, a voxel based CRF with cliques of up to size three, is used to identify candidate lesions. In the second level, higher order potentials are incorporated leveraging robust textural features which are invariant to rotation and local intensity distortions. Furthermore, we show how to exploit temporal and longitudinal images, should they be available, into the AMCRF model. The proposed algorithm is tested on 120 multimodal clinical datasets acquired from Relapsing-Remitting MS patients during multi-center clinical trials. Results show a sensitivity of 93%, a positive predictive value of 70% and average False Positive (FP) counts of 0.77. Moreover, the temporal AMCRF results show the same sensitivity as the AMCRF model while decreasing the FP counts by 22%.


Multiple Sclerosis Brain Magnetic Resonance Image False Detection Textural Pattern Voxel Level 
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.
    Miki, Y., et al.: Computer-assisted quantitation of enhancing lesions in multiple sclerosis: correlation with clinical classification. Am. J. Neur. 18, 705–710 (1997)Google Scholar
  2. 2.
    Bedell, B., Narayana, P.: Automatic segmentation of Gadolinium-enhanced multiple sclerosis lesions. Magn. Reson. Med. 39, 935–940 (1998)CrossRefGoogle Scholar
  3. 3.
    He, R., Narayana, P.: Automatic delineation of Gd enhancements on magnetic resonance images in multiple sclerosis. Med. Phys. 29, 1536–1546 (2002)CrossRefGoogle Scholar
  4. 4.
    Datta, S., et al.: Segmentation of gadolinium-enhanced lesions on MRI in multiple sclerosis. J. Magn. Reson. Imag. 25, 932–937 (2007)CrossRefGoogle Scholar
  5. 5.
    Karimaghaloo, Z., Shah, M., Francis, S.J., Arnold, D.L., Collins, D.L., Arbel, T.: Detection of gad-enhancing lesions in multiple sclerosis using conditional random fields. In: Jiang, T., Navab, N., Pluim, J.P.W., Viergever, M.A. (eds.) MICCAI 2010, Part III. LNCS, vol. 6363, pp. 41–48. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  6. 6.
    Karimaghaloo, Z., et al.: Automatic detection of Gadolinium-enhancing multiple sclerosis lesions in brain MRI using conditional random fields. TMI (2012)Google Scholar
  7. 7.
    Karimaghaloo, Z., Arnold, D.L., Collins, D.L., Arbel, T.: Hierarchical conditional random fields for detection of gad-enhancing lesions in multiple sclerosis. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012, Part II. LNCS, vol. 7511, pp. 379–386. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  8. 8.
    Lazebnik, S., Schmid, C., Ponce, J.: A sparse texture representation using local affine regions. IEEE PAMI 27, 1265–1278 (2005)CrossRefGoogle Scholar
  9. 9.
    Leo, B.: Random forests. Machine Learning, 5–32 (2001)Google Scholar
  10. 10.
    Rubner, Y., Tomasi, C., Guibas, L.: The earth mover’s distance as a metric for image retrieval. IEEE PAMI, 99–121 (2000)Google Scholar
  11. 11.
    Taskar, B., Chatalbashev, V., Koller, D., Guestrin, C.: Learning structured prediction models: a large margin approach. In: ICML, pp. 896–903 (2005)Google Scholar
  12. 12.
    Boykov, Y., Kolmogorov, V.: An experimental comparison of min- cut/max-flow algorithms for energy minimization in vision. PAMI, 1124–1137 (2004)Google Scholar
  13. 13.
    Smity, S.: Fast robust automated brain extraction. Hum. Brain Mapp., 143–155 (2002)Google Scholar
  14. 14.
    Nyul, L., Udupa, J.: On standardizing the MR image intensity scale. Comp. Assisted Tomography 42, 1072–1081 (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Zahra Karimaghaloo
    • 1
  • Hassan Rivaz
    • 2
  • Douglas L. Arnold
    • 3
  • D. Louis Collins
    • 2
  • Tal Arbel
    • 1
  1. 1.Centre for Intelligent MachinesMcGill UniversityCanada
  2. 2.Montreal Neurological InstituteMcGill UniversityCanada
  3. 3.NeuroRx ResearchMontrealCanada

Personalised recommendations