Morphological Scale-Space Decomposition for Segmenting the Ventricular Structure in Cardiac MR Images

Abstract

We propose a method for segmenting the endocardial contour of the left ventricle from magnetic resonance (MR) images, using morphological scale-space decomposition based on multiscale spatial analysis. This approach comprises a powerful tool which presents many advantages: the preservation of scale-space causality, the localisation of sharp-edges, and the reconstruction of the original image from the scale-space decomposition. An appropriate scale is defined as the scale that maximizes the response of the morphological filter through the scale-space at each point giving constant scale values in a region of constant width. The approach was able to separate the gray-level appearance structures inside the ventricular cavities from the endocardial contour, facilitating the segmentation process.