87 Thiyam Ibungomacha Singh, Romesh Laishram, Sudipta Roy, Ahanthem Romeo, Lukram Somorjit Singh, Maibam Rojit meitei International Journal of Computer & Mathematical Sciences IJCMS ISSN 2347 – 8527 Volume 6, Issue 3 March 2017 Medical Image Segmentation Using Fuzzy C Means Clustering and bat Algorithm Thiyam Ibungomacha Singh 1 ,Romesh Laishram 2 , Sudipta Roy 3 , Ahanthem Romeo 4 ,Lukram Somorjit Singh 5 ,Maibam Rojit meitei 6 1 Manipur Institute of Technology, Takyelpat, Imphal 2 Manipur Institute of Technology, Takyelpat, Imphal 3 National Institute of Technology, Silchar 4 Manipur Institute of Technology, Takyelpat, Imphal 4 Manipur Institute of Technology, Takyelpat, Imphal 5 Manipur Institute of Technology, Takyelpat, Imphal ABSTRACT In this paper we have undertaken segmentation problem of brain MRI image. The problem of segmentation is performed using the popular Fuzzy C Means (FCM) clustering algorithm. The optimization of the cluster centers of FCM is carried out based on bat algorithm which is an optimization algorithm. The proposed algorithm is called BATFCM. The simulation result shows that the hybrid algorithm performs better than FCM in producing the precise image for detecting brain anomalies which will be helpful in medical diagnosis and for further analysis. I. Introduction Medical image analysis is currently being of research interest in digital image processing domain especially Magnetic Resonance Imaging (MRI) images. The objective of applying digital image processing tools in medical images is to aid the radiologist for proper diagnosis. The field of medical image processing has seen tremendous growth in the last decade. This new research direction is a result of substantial improvement in the field of Digital Image Processing. The medical images X-Ray, ultrasound, computed tomography (CT) scan and Positron Emission Tomography (PET) gives a number of information about body structures. More advanced technology has been developed in the medical imaging such as the Magnetic Resonance Imaging (MRI). Unlike other imaging modalities, MRI gives the perfect result as well as no biological hazards. MRI operates at Radio-Frequency (RF) range; thus there are no ionizing radiations involved. Furthermore, MRI can generate excellent soft tissue contrast. The basic goal in segmentation process is to partition an image into regions that are homogeneous with respect to one or more characteristics. In medical image processing for the detection of tumors, measurement of tumor volumes and its response to therapy, detection of the coronary border in angiograms, automated classification of blood cells, detection of micro calcifications on mammograms, heart image extraction from cardiac cine angiograms etc.Segmentation tool is used [1]. Segmentation refers to the process of partitioning a digital image into multiple segments or regions. The goal of segmentation is to simplify the representation of an image into something that is more meaningful and easier to analyze. Image segmentation is typically used to locate objects and boundaries in images [2]. One natural view of segmentation [2] is that we are attempting to determine which components of a data set naturally "belong together". Clustering is a process which means a data set is replaced by clusters, which are collections of data points that "belong together". FCM is a fuzzy clustering method which was proposed by J.C. Bezdek in 1981 [3]. This is a powerful clustering technique for medical image segmentation. But the FCM clustering algorithm sometimes degrades the accuracy of the image because it takes only the pixel attributes for clustering and can only attain the local minima. To avoid this drawback, optimization algorithms are applied. The are many natured inspired optimization algorithms such as Genetic Algorithm (GA) Particle swarm optimization (PSO) and Artificial Bee Colony Algorithm (ABC). These algorithms are quite popular but in this paper we have introduced the application of bat algorithm with FCM for segmentation problem II. Fuzzy C Means Algorithm Segmentation is greatly being improved by using the fuzzy C-means (FCM) algorithm instead of using K- Means Clustering algorithm. K-Means Clustering algorithm is also known as the Hard C-means algorithm where each data point is a member of one and only one cluster and it has well defined boundary between