Medical Image Segmentation with active contour and optimization Techniques: Survey

Abstract

The principal aim is to improve medical diagnosis through segmented images. Thus, medical image segmentation become one of the key technologies in computer-aided diagnosis. Active contours, or snakes, have been widely used for image segmentation purposes. However, high noise sensitivity and poor performance over weak edges are the most acute issues that hinder the segmentation accuracy of these curves, particularly in medical images. To overcome these issues, a novel external force that integrates gradient vector flow (GVF) field forces and the traditional snake function is proposed in this research. In addition, a novel technique is applied to limit the boundary of the initial contour by set four initial points around the medical issue and then connecting them by polynomial curves. Moreover, the positive effect of Particle Swarm Optimization (PSO) on calculating the final active contour area and its percentage to the entire image area is proved in this work.