• DocumentCode
    2359753
  • Title

    Quantification of myocardial edema and necrosis during acute myocardial infarction

  • Author

    Baron, N. ; Kachenoura, Nadjia ; Beygui, F. ; Cluze, P. ; Grenier, P. ; Herment, A. ; Frouin, F.

  • fYear
    2008
  • fDate
    14-17 Sept. 2008
  • Firstpage
    781
  • Lastpage
    784
  • Abstract
    Quantification of myocardial edema and necrosis during acute myocardial infarct (MI) is crucial for patientpsilas prognosis. The aim of this study was to evaluate these two parameters from MRI late gadolinium enhancement (LGE) and T2 weighted short Tau inversion recovery (STIR) black-blood sequences acquired in 22 patients with MI. To estimate the necrosis and edema volumes, a clustering method based on a fuzzy c-means algorithm was used. Results were compared against a manual delimitation and a semi-automatic thresholding currently reported in the literature. The proposed estimation of the necrosis volume was strongly correlated with both the semi automatic method and the manual delineation (r>0.9). For the quantification of the edema, the approach was valid, except for small size infarcts. Thus the automated quantification of necrosis is reliable compared to conventional approaches, and the method is encouraging for the quantification of edema.
  • Keywords
    biomedical MRI; blood; cardiology; fuzzy set theory; image enhancement; image sequences; medical image processing; pattern clustering; MRI late gadolinium enhancement; acute myocardial infarction; automated necrosis quantification; black-blood sequences; clustering method; fuzzy c-means algorithm; myocardial edema quantification; myocardial necrosis quantification; necrosis volume estimation; patient prognosis; weighted short Tau inversion recovery; Anatomy; Cardiology; Clustering algorithms; Clustering methods; Data visualization; Image segmentation; Magnetic resonance imaging; Myocardium; Phased arrays; Radiology;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computers in Cardiology, 2008
  • Conference_Location
    Bologna
  • ISSN
    0276-6547
  • Print_ISBN
    978-1-4244-3706-1
  • Type

    conf

  • DOI
    10.1109/CIC.2008.4749158
  • Filename
    4749158