Techniques for cardiac image segmentation

Abstract

Two algorithms for medical image processing are discussed: CAVITY DETECTOR, which solves the segmentation problem of regions which are not completely surrounded by walls and EDGMENTATION, which is used for (i) preprocessing before segmentation, (ii) boundary refinement before edge detection, (iii) segmentation based on a modified split-and-merge approach, and (iiii) data reduction. In this paper, we describe the principles of the algorithms and illustrate their generic properties by discussing the results of applications in 2-D and 3-D cardiac MRI and 3-D and 4-D cardiac SPECT.