Brain Tumor Detection in MRI Images, using Fuzzy C-Means and Cuckoo Search Algorithm

Abstract

: Image segmentation is an important technique in computer vision and image processing applications. Magnetic Resonance Image (MRI) is the key process, in detecting the brain tumor. Fuzzy C-Means (FCM) algorithm has been widely used for medical segmentation. But the major drawback of the FCM is to consume more computational time to convergence. The effectiveness of the algorithm in terms of improves the computational rate by optimizing the cluster center. A new hybrid technique is proposed in this paper fuzzy c-means combining cuckoo search algorithm and Support Vector Machine (SVM) for brain tumor detection. Image noise is removed by adaptive median filter, enhanced the image using Gaussian stretch and skull stripping is done by morphological operations. Fuzzy c means (FCM) clustering is used for segmentation of the image to observe the suspicious region in brain MRI image. Cuckoo search algorithm (CSA) is used to optimize cluster center of FCM of the brain image, after which SVM method is applied to classify the brain MRI images, which provide accurate result for classification of brain MRI images.

Keywords - Image segmentation, Magnetic Resonance Image (MRI), Fuzzy C-Means(FCM), Cuckoo Search Algorithm (CSA), Support Vector Machine (SVM)