• DocumentCode
    2717632
  • Title

    A high-end binary search unit for svm dna micro-arrays

  • Author

    Sotiropoulos, Y. ; Sotiriades, E. ; Papaefstathiou, I.

  • Author_Institution
    Dept. of Electron. & Comput. Eng., Tech. Univ. of Crete, Chania
  • fYear
    2008
  • fDate
    16-18 Dec. 2008
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    The Use of DNA micro-arrays helps scientist detecting several types of cancer at primitive stage. There is a great amount of data that have to be processed when using DNA micro-arrays by computational heavy algorithms. For these two reasons, the great scientific interest of the problem, and the heavy computational load of the algorithm, bioinformatics research community have proposed several approaches, while the reconfigurable logic research community has implement certain systems for performance boosting. At this paper we have profiled the algorithm that produce the most accurate results, and implement the computational intensive functions in a reconfigurable based system. This approach produces up to 500 times performance boosting in comparison with a conventional general purpose processor using Matlab.
  • Keywords
    biocomputing; bioinformatics; mathematics computing; support vector machines; Matlab; SVM DNA micro-arrays; bioinformatics research community; computational heavy algorithms; high-end binary search unit; support vector machine; Bioinformatics; Boosting; Cancer detection; DNA computing; Delay; Laboratories; Mathematical model; Reconfigurable logic; Signal processing algorithms; Support vector machines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Innovations in Information Technology, 2008. IIT 2008. International Conference on
  • Conference_Location
    Al Ain
  • Print_ISBN
    978-1-4244-3396-4
  • Type

    conf

  • DOI
    10.1109/INNOVATIONS.2008.4781700
  • Filename
    4781700