A hybrid harmony search algorithm for MRI brain segmentation

Abstract

Automatic magnetic resonance imaging (MRI) brain segmentation is a challenging problem that has received significant attention in the field of medical image processing. In this paper, we present a new dynamic clustering algorithm based on the hybridization of harmony search (HS) and fuzzy c-means to automatically segment MRI brain images in an intelligent manner. In our algorithm, the capability of standard HS is modified to automatically evolve the appropriate number of clusters as well as the locations of cluster centers. By incorporating the concept of variable length encoding in each harmony memory vector, this algorithm is able to represent variable numbers of candidate cluster centers at each iteration. A new HS operator, called the ''empty operator'', has been introduced to support the selection of empty decision variables in the harmony memory vector. The PBMF cluster validity index is used as an objective function to validate the clustering result obtained from each harmony memory vector. Evaluation of the proposed algorithm has been performed using both real MRI data obtained from the Center for Morphometric Analysis at Massachusetts General Hospital and simulated MRI data generated using the McGill University BrainWeb MRI simulator. Experimental results show the ability of this algorithm to find the appropriate number of naturally occurring regions in brain images. Furthermore, the superiority of the proposed algorithm over various state-of-the-art segmentation algorithms is demonstrated quantitatively.

A New Image Clustering Method Based on the Fuzzy Harmony Search Algorithm and Fourier Transform

“…(12), was proposed by Fukuyama and Sugeno in 1989. Where, m=1, xk is the kth data point, vi are cluster prototypes (cluster centers), c is the number of clusters, v is the grand mean of all data xk , uik is the membership value of data xk of class ci, and |ci| is the total amount of data belonging to cluster i.…”

“…This situation illustrates the imprecision and uncertainty related to the membership of the object to more than one group. The solution to this problem lies in the introduction of fuzzy logic, by adding a membership value that indicates the degree of membership of an element with different groups [12][13][14]. (1)…”

Harmony Search (HS) Algorithm for Solving Optimal Reactive Power Dispatch Problem

“…Harmony search (HS) Geem et al [10], [11] is a relatively new population-based metaheuristic optimization algorithm, that imitates the music improvisation process where the musicians improvise their instruments' pitch by searching for a perfect state of harmony. It was able to attract many researchers to develop HS-based solutions for many optimization [12]- [16] problems such as music composition ,ground water modeling (Ayvaz 2007(Ayvaz , 2009 [17]- [23]. HS imitates the natural phenomenon of musicians' behavior when they cooperate the pitches of their instruments together to achieve a fantastic harmony as measured by aesthetic standards.…”

Brain MRI Segmentation based on Different Clustering Algorithms

“…In this study, especially brain images were segmented. The CSM (Cohesion based Self-Merging)-based partial k-means clustering algorithm [3] was used for the segmentation in brain images in the studies carried out, FCM (Fuzzy C-Means) and artificial neural networks [4], flock-based clustering algorithm [5], FCM-based HS (Harmony Search) hybrid algorithm [6], and fuzzy-logic based genetic algorithm [7] were used in order to increase clustering success. After using K-means and SVM (Support Vector Machine) [8] in order to overcome the difficulties that arise as a result of the failure to find the noise and reference image in MR images, and wavelet transform to reduce the noise in the images, the K-means clustering algorithm [9] was used for segmentation.…”