Brain tumor is a complex disease that occurs due to the abnormal growth of brain cells. For efficient treatment planning, earlier detection of tumor is necessary. Magnetic Resonance Imaging (MRI) is now recognized as an important tool for the detection of Brain tumor. MRI can be used to identify various tissues inside the brain with good efficiency and accuracy. The radiation used in MRI is non-ionizing and the contrast agents used are less harmful. Computer Aided Diagnosis (CAD) could be almost as effective ad double reading by providing a second opinion to the radiologist, and help in increasing the sensitivity and accuracy of detection. The proposed Brain Tumor detection algorithm is composed of four stages: Preprocessing, Segmentation, Feature extraction and Classification. Major steps in preprocessing are anisotropic filter, contrast enhancement. A novel algorithm for brain MRI segmentation using Bounding Box is proposed in this work. Watershed operation was implemented to authenticate the performance of proposed method. The parameters used to evaluate the performance of segmentation algorithm are Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR), accuracy. Gray Level Co-Occurrence Matrix (GLCM) is used to extract nine Haralick Features from brain MR images. These features vectors are used to train the Multiclass SVM algorithms. Finally all these image processing steps are integrated and get the results.
Volume 11 | 07-Special Issue
Pages: 1033-1042