Performance Assessment of k-Means, FCM, ARKFCM and PSO Segmentation Algorithms for MR Brain Tumour Images

A brain tumour is a group of abnormal cells that grows in or around the brain. Brain tumours are either malignant or benign. A malignant tumour, also called brain cancer, grows rapidly and often invades healthy areas of the brain. Benign brain tumours do not contain cancer cells and are usually slow growing.Currently, Magnetic Resonance Imaging (MRI) is the most sensitive imaging test of the head, particularlythe brain in routine clinical practice. In order to visualise anatomic structures (tissues, body organs and nodules) of interest from medical images, segmentation plays an indispensable role. Because of theirunpredictable appearance and shape, segmenting braintumours from MR images is one of the most challenging tasks in medical image analysis. The aim of thisstudy is to compare the performance of k-means, Fuzzy C-Means (FCM), Particle Swarm Optimisation (PSO) and Adaptive Regularised Kernel Fuzzy C-Means (ARKFCM)-based segmentation techniques for accurate delineation of tumour using clinical brain tumour MR images. Preprocessing is carried out to remove noise and to achieve better segmentation results. The performance of these segmentation algorithms is evaluated using Peak Signalto-Noise Ratio (PSNR), Mean Square Error (MSE), Normalised Cross Correlation (NCC), Structural Similarity Index (SSIM) and segmentation accuracy. Experimental evaluation revealed k-means and FCM segmentation algorithms out performed compared with PSO and ARKFCM segmentation algorithms.