3-D Medical Image Region-Growing based Segmentation Techniques, Challenges and Open Issues

Abstract

Volumetric medical imaging acquisition technologies such as Computed Tomography (CT), Magnetic Resonance Tomography (MRT) and Positron Emission Tomography (PET) provide an effective means for noninvasive mapping of the anatomy of a subject. With these technologies being used every day there are enormous number of medical images, this has necessitated the use of computers in processing and analysis of these images. A significant task in medical image analysis is segmentation, whose goal is to partition a volumetric medical image into separate regions, usually anatomic structure (tissue type) that are meaningful for a specific task. In this paper region growing based segmentation techniques are discussed and the generic algorithms are given. The challenges and open issues in the medical field of this type of techniques are also highlighted. A recommendation of how the techniques can be used in the medical field is proposed.