Identificationc and Classification of Benign and Malignant Tumors in Brain using k-NN and NN intelligent algorithms K. Sudharani, Associate Professor EIE Dept, VNR Vignana Jyothi IET, sudharani_k@vnrvjiet.in Dr.T.C. Sarma 2 Former Deputy Director, NRSA,Hyderabad Dr. K. Satya Prasad 3 Professor ECE dept, JNTU, Kakinada Abstract- In brain tumor diagnosis and radiotherapy planning Magnetic resonance imaging (MRI) plays a very important role for tumor segmentation. Due to the great diversity in appearance of tumor tissue among different patients and the unclear boundaries of lesions to Automating this process is a challenging task. This paper describes an automatic algorithm to differentiating Benign and Malignant tissues. The traditional techniques for CT and MRI images classification and tumor detection is by human examination. The classification techniques inspected by the operator are unrealistic for huge amounts of data and are also non reproducible. A noise caused in CT and MRI images by operator which can lead to severe inaccuracies in classification. The proposed method is created in the following computational technique; such as MRI image acquisition, Color plane extraction, re sampling of the image, classification of the tumor and cancer, training and testing. The concept of Gray level co-occurrence matrix (GLCM) is used in feature extraction. Quantitative and qualitative analysis is conducted on 50 MRI and 10 ground truth images of the disordered brain for the validation performance of the proposed brain tissue classification technique. Keywords: Benign, Malignant, Texture Analysis, GLCM I. INTRODUCTION A brain tumor is a growth of abnormal cells in or close to the brain. There are several types of brain tumors, when diagnosed there are to be distinguished as follows: 1. The type and grade 2. in case it is primary or secondary tumor 3. In case it is cancerous or non-cancerous 4. Tumor location. Malignant tumors hold cancer cells and often do not have clear borders. These rarely spread beyond the brain or spinal cord. Benign tumors doesn’t contain any cancer cells. They grow slowly and can often be removed. Whether cancerous or not, tumors [8] that origin in the cells of the brain are called Primary brain tumors. Secondary brain tumors begin in other portions of the body and then spread into the brain. Grading characterizes how fast the cells can cultivate and are likely to spread. They are classified as: Lower grade tumors which are not aggressive and are usually with long-time survival. Higher grade tumors can grow quickly and cause damage which are difficult to treat. Tumors can have various grades of cells, though the most malignant cell regulates the grade for the entire tumor. Some tumors can alter the way they grow and develop malignant over time. MRI [9] (Magnetic resonance imaging) helps in obtaining a structural three-dimensional images of various regions of the human body. Cancer plays a crucial role in tumor formation inside the human body. Manual segmentation and tumor detection using fuzzy k-means [1] algorithm provided necessary tumor identification. Undergoing radiotherapy [2] for primary brain tumors research about the automatic segmentation on MRI brain images which proved acceptable accuracy and reproducibility. While some tasks are partially automated for tumor detection, some manual interface is still required and this manual work is time-consuming task. For this, automatic detection for tumor in multi-series MRI analysis [3] research is performed. International Journal of Computer Science and Information Security (IJCSIS), Vol. 14, No. 10, October 2016 790 https://sites.google.com/site/ijcsis/ ISSN 1947-5500