• DocumentCode
    1788182
  • Title

    MS lesions segmentation in 3D MR images using FCM and SVM

  • Author

    Merzoug, Amina ; Benamrane, Nacera ; Ahmed, Abdelmalik Taleb

  • Author_Institution
    Dept. of Comput. Sci., USTO-MB, Oran, Algeria
  • fYear
    2014
  • fDate
    14-17 Oct. 2014
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    This paper proposes an approach to automatically segment MS lesions in MR images using fuzzy c-means (FCM) and a support vector machines (SVM) based on the sequential minimal optimization (SMO) in learning step. A postprocessing based on morphological operations was applied to refine the obtained results. The proposed approach was tested on 3D MR images and the obtained results are encouraging.
  • Keywords
    biomedical MRI; fuzzy set theory; image segmentation; medical image processing; optimisation; support vector machines; 3D MR images; FCM; MS lesion segmentation; SVM; fuzzy c-means; morphological operations; sequential minimal optimization; support vector machines; Image segmentation; Lesions; Magnetic resonance imaging; Multiple sclerosis; Optimization; Support vector machines; Three-dimensional displays; 3D MR image; FCM; SMO; SVM; multiple sclerosis; segmentation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing Theory, Tools and Applications (IPTA), 2014 4th International Conference on
  • Conference_Location
    Paris
  • Print_ISBN
    978-1-4799-6462-8
  • Type

    conf

  • DOI
    10.1109/IPTA.2014.7001924
  • Filename
    7001924