Automatic segmentation of ventricular endocardium on cine-CT images by an active contour process

Author

Sebbahi, Ali ; Herment, Alain ; Cesare, Alain De

Author_Institution

Hopital Broussais, Inst. Nat. de la Sante et de la Recherche Med., Paris, France

fYear

1994

fDate

3-6 Nov 1994

Firstpage

522

Abstract

An automatic detection method has been developed for cine-CT images. A preprocessing relying on histogram thresholding is first applied, permitting the extraction of an initial endocardial contour. An iterative processing based on an active contour algorithm and a spline optimization model allows the deformation of the initial contour to an optimal endocardial contour which takes into account specific information about papillary muscles

Keywords

cardiology; computerised tomography; image segmentation; iterative methods; medical image processing; optimisation; active contour algorithm; active contour process; automatic segmentation; cine-CT images; histogram thresholding; initial contour deformation; iterative processing; medical diagnostic imaging; optimal endocardial contour; papillary muscles; spline optimization model; ventricular endocardium; Active contours; Computed tomography; Data mining; Deformable models; Heart; Histograms; Image edge detection; Image segmentation; Interpolation; Iterative algorithms; Muscles; Spline;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1994. Engineering Advances: New Opportunities for Biomedical Engineers. Proceedings of the 16th Annual International Conference of the IEEE

Conference_Location

Baltimore, MD

Print_ISBN

0-7803-2050-6

Type

conf

DOI

10.1109/IEMBS.1994.411923

Filename

411923