• DocumentCode
    320134
  • Title

    Detection of MRI brain contour using nonlinear anisotropic diffusion filter

  • Author

    Gallo, G. ; Zingale, A. ; Zingale, R.

  • Author_Institution
    Dipartimento di Matematica, Catania Univ., Italy
  • Volume
    3
  • fYear
    1996
  • fDate
    31 Oct-3 Nov 1996
  • Firstpage
    1062
  • Abstract
    Automated boundary extraction techniques applied to detect MRI brain contour do not provide, until now, contours as accurate as required in critical medical applications; hence contouring is actually accomplished by manual or interactive edge detection methods. In order to overcome such inadequate results, scale-space technique using anisotropic diffusion filter has been applied. After smoothing the gradient has been computed using the nonmaxima suppression technique. When the hysteresis thresholding has been performed the brain contour appeared thinned and almost entirely connected. The used technique provided a great advantage because it allowed to detect the true location and morphology of sulci and gyri
  • Keywords
    biomedical NMR; brain; edge detection; medical image processing; MRI brain contour detection; anisotropic diffusion filter; automated boundary extraction techniques; critical medical applications; gyri; hysteresis thresholding; interactive edge detection; magnetic resonance imaging; medical diagnostic imaging; nonlinear anisotropic diffusion filter; nonmaxima suppression technique; scale-space technique; smoothing; sulci; Anisotropic magnetoresistance; Biomedical equipment; Filters; Hysteresis; Image edge detection; Magnetic resonance imaging; Manuals; Medical services; Morphology; Smoothing methods;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE
  • Conference_Location
    Amsterdam
  • Print_ISBN
    0-7803-3811-1
  • Type

    conf

  • DOI
    10.1109/IEMBS.1996.652709
  • Filename
    652709