Bhawna Gupta et al, International Journal of Computer Science and Mobile Computing, Vol.3 Issue.4, April- 2014, pg. 1259-1264 © 2014, IJCSMC All Rights Reserved 1259 Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology ISSN 2320–088X IJCSMC, Vol. 3, Issue. 4, April 2014, pg.1259 – 1264 RESEARCH ARTICLE Brain Tumor Detection using Curvelet Transform and Support Vector Machine Bhawna Gupta, Shamik Tiwari CSE, Mody University of Technology & Science, India Abstract — The prevalent cause of death in human being is brain tumor. A brain tumor is a mass or growth of anomalous cells in brain. The detection of brain tumor is difficult task. Image processing provides relevant techniques for efficient detection. In the proposed technique, first the features of MRI (Magnetic Resonance Imaging) images are extracted with curvelet transform, and then these features are applied to the support vector machine for successful identification. This proposed methodology gives efficient results. Keywords— Brain Tumor; Curvelet Transform; Support Vector Machine I. INTRODUCTION A brain tumor is a mass of irrelevant cells growing in the brain or central spine canal. There are two basic kinds of brain tumors – primary brain tumors (benign) and metastatic (malignant) brain tumors. Primary brain tumors initiate and stay in the brain. Metastatic brain tumors begin as cancer somewhere in the body and spread to the brain [1]. MRI (magnetic resonance imaging), CT (computed tomography) are most adequate way to locate the brain tumor. The typical methods, which are present in diagnosis, are human inspection, biopsy, expert opinion and etc. These methods have some drawbacks like biopsy take so much time and human inspection is not always correct [2]. So image processing techniques are used to identify brain tumor. A number of studies have been done on brain tumor detection. Lahmiri et al. [3] designed an automated system in which it performs lobe asymmetry to distinguish the normal and abnormal brain MRI images. The proposed automated diagnosis system includes four 1). The original image is processed with a Laplacian of Gaussian (log) filter for noise filtering and edge enhancement. 2) the filtered image is split into two right and left lobe sub-images 3) the relevant features are extracted to account for asymmetry 4) the resulting feature vector feeds the input of a support vector machine (svm). Gupta et al. [4] proposed a methodology in which image processed through histogram equalization, binarization, morphological operations, region isolation, feature extraction and neuro classifier. Gray level co-occurrence matrix is used for feature extraction and neural network is used for classification. Al-badarneh et al. [5] proposed an approach which have following steps feature extraction and classification. In feature extraction main texture feature are extracted. Then neural network and k-nearest neighbour is used for classification. Gupta et al. [6] designed a system in which feature extraction is done using curvelet transform on the lung cancer CT scans. The rest of paper is organized as follows: Section II describes the proposed methodology. Section III shows experimental results. Section IV presents conclusion for the proposed method.