An efficient method for brain tumor detection and categorization using MRI images by K-means clustering & DWT

Abstract

Brain tumor is an uncontrolled mass of tissues in the brain which originate due to mutated growth of tissues. Brain tumor has become a leading cost of death in modern day environment and researchers are inclined to find ways to mitigate the proliferation of this disease. A lot of methods have been applied in brain tumor detection ranging from image processing to signal based analysis. In this study a robust image processing based method is applied using MRI images. MRI images are preferred due to their simplicity and low noise presence. In this study first a clustering based method is used to segment the image and then SVM is applied for tumor detection. A total seven features were considered and were analyzed by the classifiers. SVM with 94.6% accuracy gave a robust result.