Deformable contour-based method for medical image segmentation
Domagoj Kovačević, Sven Lončarić · 2002
: In this paper we present a method for segmentation of medical images based on a deformable contour paradigm. The deformable contour is a novel approach in image segmentation and has some advantages over classical segmentation methods such as region basedoredge mased methods. We are using the level set inplementation of the deformable contour paradigm. The graphical user interface for the algorithm is also developed. Images we are segmenting are computed tomography #CT# scans of the abdominal region of the human body and our region of interest is aorta #blood vessel#. For one of the volumes results are visualized in 3-D. Keywords: image segmentation, active contour, computed tomography 1. Introduction In this work we present a method for segmentation of computer tomography #CT# images of the human abdominal region. Typical CT image is shown in Figure 1. Data set for each patient consists of forty to #fty such images taken with di#erent values of height coordinate so all images toge...