Segmenting MRI brain images using evolutionary computation technique

Medical image segmentation is a fundamental preprocessing step in most systems that supports diagnosis or planning of surgical operations. The traditional Fuzzy c means clustering algorithm performs well in the absence of noise. Traditional FCM leads to its non robust result mainly due to 1. Not utilizing the spatial information in the image. 2. Use of Euclidean distance. These limitations can be addressed by using robust spatial kernel FCM (RSKFCM). RSKFCM consider the spatial information and uses Gaussian kernel function to calculate the distance between the center and data points. Though RSKFCM gives good result, the main drawback behind this method is the inability of generating global minima for the objective function. To improve the efficiency of RSKFCM method, in this paper we proposed the genetic algorithm based RSKFCM. By using the genetic algorithm, RSKFCM initializes the cluster centers and reaches the global minima of the objective function. Experimentation is carried out on the standard brain image dataset. Experimental result reveals that the proposed genetic algorithm based RSKFCM outperforms other FCM methods with the use of various cluster validity functions.