Immune system programming for medical image segmentation.

This paper introduces an automatic strategy for the segmentation of medical images from Magnetic Resonance Imaging (MRI) and Computed Topography (CT). A new segmentation technique is proposed to combine a new evolutionary algorithm, called the Immune System Programming (ISP) algorithm, with the Region Growing (RG) technique. The ISP algorithm with a tree data structure has the ability to create new mathematical threshold functions, and RG can use these functions to achieve an efficient segmentation process for medical images. Several MRI images with different levels of Radio Frequency (RF) and noise are used to test the proposed segmentation technique. In different experiments, the proposed technique showed promising performance and produced a new set of efficient threshold functions.