An effective method for segmentation of MR brain images using the ant colony optimization algorithm

Abstract

Since segmentation of magnetic resonance images is one of the most important initial steps in brain magnetic resonance image processing, success in this part has a great influence on the quality of outcomes of subsequent steps. In the past few decades, numerous methods have been introduced for classification of such images, but typically they perform well only on a specific subset of images, do not generalize well to other image sets, and have poor computational performance. In this study, we provided a method for segmentation of magnetic resonance images of the brain that despite its simplicity has a high accuracy. We compare the performance of our proposed algorithm with similar evolutionary algorithms on a pixel-by-pixel basis. Our algorithm is tested across varying sets of magnetic resonance images and demonstrates high speed and accuracy. It should be noted that in initial steps, the algorithm is computationally intensive requiring a large number of calculations; however, in subsequent steps of the search process, the number is reduced with the segmentation focused only in the target area.

Fig. 1

Results of brain segmentation that are from left to right, original image, divided to GM, WM, and CSF via the proposed ACO algorithm

Fig. 2

Results of brain segmentation that extract WM removing gray matter. Figures from left to right are the original image, which results after GA, PSO, and ACO

Fig. 3

Results of brain segmentation that extract gray matter removing white matter. Figures, from left to right, consist of original image, which results after GA, PSO, and ACO

Fig. 4

Results of each algorithm segmenting the brain. This figure, from left to right consists of original image, which results after GA, PSO, and ACO

Fig. 5

Magnification of key areas in Fig. 2 demonstrating the differences in the performance of GA, PSO, and ACO

Fig. 6

Magnification of key areas in Fig. 3 demonstrating the differences in the performance of GA, PSO, and ACO

Fig. 7

Convergence characteristic of segmentation algorithms; error against FEs, for images in Fig. 2

Fig. 8

Convergence characteristic of segmentation algorithms; error against FEs, for images in Fig. 3

Fig. 9

Convergence characteristic of segmentation algorithms; error against FEs, for images in Fig. 4