A Multi-agent System Approach for Medical Image Segmentation

Abstract

Image segmentation still requires improvements although there have been research works since the last few decades. This is coming due to some issues. Firstly, most image segmentation solutions are problem-based. Secondly, medical image segmentation methods generally have restrictions because medical images have very similar gray level and texture among the interested objects. The goal of this work is to design a framework to extract simultaneously several objects of interest from Computed Tomography (CT) images by using some priori-knowledge. Our method used properties of agent in a multi-agent environment. The input image is divided into several sub-images, and each agent works on a sub-image and tries to mark each pixel as a specific region by means of given priori-knowledge. During this time the local agent marks each cell of sub-image individually. Moderator agent checks the outcome of all agents’ work to produce final segmented image. The experimental results for cranial CT images demonstrated segmentation accuracy around 90%.