ISSN: 2277-9655 [Babu* et al., 6(7): July, 2017] Impact Factor: 4.116 IC™ Value: 3.00 CODEN: IJESS7 http: // www.ijesrt.com © International Journal of Engineering Sciences & Research Technology [838] IJESRT INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY A NOVEL APPROACH TO BRAIN TUMOR DETECTION B. Shoban Babu *1 , S. Varadarajan 2 *1 Associate Professor, Department of Electronics & Communication Engineering, Sri Venkateswara College of Engineering and Technology , R. V. S. Nagar, Chittoor - 517127, Andhra Pradesh, India 2 Professor, Department of Electronics & Communication Engineering, S. V. U College of Engineering, Sri Venkateswara University, Tirupati – 517502, Andhra Pradesh, India DOI: 10.5281/zenodo.834556 ABSTRACT An intracranial mass of abnormal cells in the brain that have grown out of control is referred to as brain tumor. Based on the type of tissue involved and the location of the tumor, brain tumors are classified as benign (non- cancerous) or malignant (cancerous). Brain tumor prognosis attributed lot of significance in successful treatment. Magnetic Resonance Imaging (MRI) is proved to be a most accurate diagnostic tool for human soft tissue analysis. However brain tumor segmentation and classification is a cumbersome process, as magnetic resonance images are inherently noisy in nature. In this work Support Vector Machine (SVM) classifier is used to classify the MRI images as normal and abnormal (tumor). Features are extracted from the segmented images and the clustered to improve the SVM classifier accuracy. Statistical analysis is performed with 10 fold cross validation to find the robustness of the classifier. Experimental result shows 96.5 percent accuracy while testing with MRI brain tumor images from IMRI Volumetric Non-Rigid Registration-Dataset. KEYWORDS: Magnetic Resonance Imaging, Support Vector Machine, Brain Tumor, Statistical analysis, Cross Validation. I. INTRODUCTION Uncontrolled proliferation of abnormal cells is referred to as brain tumor. Brain tumors are termed as primary, if those originate in the brain. The secondary tumors originate at other body parts and spread to the brain through metastasis. Primary brain tumors can be benign (non-cancerous) or malignant (cancerous), whereas secondary brain tumors are malignant. Both primary and secondary tumors are dangerous and could prove fatal if not detected in early stage. As the space inside the skull is very limited, the growth of tumor inside the skull could increase the intracranial pressure causing edema, reduced blood flow, displacement, and degeneration of other tissues that control important body functions. Survival rates of the affected individual with brain tumors vary widely, depending on the type of tumor [2], however for brain tumor prognosis and successful treatment therapy planning, early and accurate tumor diagnosis is imperative. Medical imaging is becoming a very important aspect in clinical applications from diagnosis to treatment [1]. Magnetic Resonance Imaging (MRI) is a very important diagnostic tool for analysis of human soft tissue. Magnetic resonance imaging is popularly used in diagnosis of brain tumor because, it is a non-invasive technique. However the diagnosis is inherently challenging due to the large variance and complexity of tumor characterization in images, such as size, structure, location and intensities and can only be performed by professional radiologists. Also Magnetic resonance images contain noise caused by operator performance that could pave path to inaccuracies in classification. Nowadays use of computers in clinical diagnosis is extensive and spread across a wide range of medical applications such as cancer prognosis, artery thickening, brain tumors etc. Magnetic resonance imaging is the ultimate option available at the present for study of tumors as it efficiently finds the tumor types, size and the tumor location. In magnetic resonance imaging method, the image is formed by magnetic field, radio waves and a computer and there is no known significant side effects for exposure of radiation. Magnetic resonance imaging technique reveals greater details of soft human tissues. Brain tumor is often symptomatic, however in absence of symptoms, in some cases brain tumor is revealed while scan is performed for some other cause. Researchers in the recent past have put-forth several ideas for successful diagnosis of brain tumor, some of the relevant ones are reviewed here.