• DocumentCode
    3645849
  • Title

    Fractal feature-based color image segmentation for a healthcare application in dermatology

  • Author

    Alexandra Căliman;Mihai Ivanovici;Noêl Richard

  • Author_Institution
    MIV Imaging Venture Laboratory, Department of Electronics and Computers, Transilvania University, Braş
  • fYear
    2011
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    We propose a computer-based e-health system in dermatology, particularly adapted for psoriasis management, in order to assist the dermatologist to objectively evaluate the severity of skin lesions and to choose the adequate treatment. The system comprises a digital color camera, a personal computer and an image processing software application. In order to assess the severity of psoriatic lesions we use the unanimously-accepted PASI score, which is a subjective score computed by the dermatologist based on the following characteristics of the affected skin regions: area, erythema, induration and scaliness. Therefore, the precise and objective measurement of the lesion area is extremely important. However, this step has to be preceded by a segmentation operation, therefore we propose a color image segmentation approach using fractal features for the local characterization of the color texture. We show our results, then we conclude this paper.
  • Keywords
    "Fractals","Image segmentation","Lesions","Image color analysis","Skin","Complexity theory","Color"
  • Publisher
    ieee
  • Conference_Titel
    E-Health and Bioengineering Conference (EHB), 2011
  • Print_ISBN
    978-1-4577-0292-1
  • Type

    conf

  • Filename
    6150374