• DocumentCode
    3562287
  • Title

    Defining angular and radial positions and parameters for myocardial pixels in cardiac MR images

  • Author

    Engan, Kjersti ; Woie, Leik ; Eftestol, Trygve

  • Author_Institution
    Dept. of Electr. Eng. & Comp. Sci., Univ. of Stavanger, Stavanger, Norway
  • fYear
    2014
  • Firstpage
    941
  • Lastpage
    944
  • Abstract
    In this work we aimed to automatically produce a measure for the angular and radial position of all pixels within the myocardium in CMR images, left ventricle, short axis view. A reference axis is chosen in an anatomically meaningful way, and this makes the angular position measure easy to relate to the American Heart Association (AHA) 17 segment model. The angular and the radial position give values for each pixel so that each pixel can be represented by a 3D vector of information [intensity value, angular position, radial position]. Using this 3D representation of each pixel, interesting parameters can easily be defined. For example looking at LGE-CMR images for patients with myocardial scar, parameters for describing the localization of the scar segments can be found automatically and objectively, and can be used for further classification of patients.
  • Keywords
    biomedical MRI; blood vessels; cardiology; image representation; image segmentation; medical image processing; 3D representation; 3D vector; AHA; American Heart Association segment model; LGE-CMR images; angular positions; cardiac MR images; information; intensity value; left ventricle; myocardial pixels; myocardial scar; myocardium; patient classification; radial positions; reference axis; scar segment localization; short axis view; Biomedical imaging; Heart; Image segmentation; Magnetic resonance imaging; Myocardium; Phase change materials;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing in Cardiology Conference (CinC), 2014
  • ISSN
    2325-8861
  • Print_ISBN
    978-1-4799-4346-3
  • Type

    conf

  • Filename
    7043199