• DocumentCode
    2394011
  • Title

    A new IC designed inside capsule endoscope for detection of bleeding region

  • Author

    Daryanavard, H. ; Karimian, G. ; Shahshahani, S.M.R.

  • Author_Institution
    Fac. of Electr. & Comput. Eng., Univ. of Tabriz, Tabriz, Iran
  • fYear
    2010
  • fDate
    3-4 Nov. 2010
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    In this paper, a new IC has been proposed for detection of bleeding region in the GI tract, to be implemented inside the capsule endoscope. In this design, firstly, by combining the image characteristics of RGB and HSI color image spaces as well as making the algorithm adaptive to image luminance, the most optimal algorithm considering the bleeding detection evaluation parameters, have been proposed; the results have shown a 5% improvement in comparison with the most efficient existing work. In contrast with the previous works, the proposed algorithm has been developed as a hardware implementation using 7-stage pipeline architecture and the input image format of Bayer CFA. The operational frequency of this circuit is 250MHz that can process each Bayer format endoscope image in less than 60μs. The whole architecture has been described by VerilogHDL and synthesized on ALTERA Stratix II FPGA.
  • Keywords
    biomedical engineering; endoscopes; field programmable gate arrays; integrated circuit design; medical image processing; pipeline processing; wounds; 7-stage pipeline architecture; ALTERA Stratix II FPGA; Bayer CFA input image format; GI tract; HSI color image spaces; IC design; RGB color image spaces; VerilogHDL; adaptive algorithm; bleeding region detection; capsule endoscope; frequency 250 MHz; hardware implementation; image luminance; optimal algorithm; Accuracy; Algorithm design and analysis; Artificial neural networks; Image segmentation; Integrated circuits; Adaptive detection; Bayer Format; Bleeding Detection; Endoscope Capsule; FPGA;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Biomedical Engineering (ICBME), 2010 17th Iranian Conference of
  • Conference_Location
    Isfahan
  • Print_ISBN
    978-1-4244-7483-7
  • Type

    conf

  • DOI
    10.1109/ICBME.2010.5704957
  • Filename
    5704957