A distinctive approach in brain tumor detection and classification using MRI

Abstract A very exigent task for radiologists is early brain tumor detection. Brain tumor raises very fast, its average size doubles in just twenty-five days. If not treated properly, the survival rate of the patient is normally not more than half a year. It can rapidly lead to death. For this reason, an automatic system is required for brain tumor detection at an early stage. In this paper, an automated method is proposed to easily differentiate between cancerous and non-cancerous Magnetic Resonance Imaging (MRI) of the brain. Different techniques have been applied for the segmentation of candidate lesion. Then a features set is chosen for every applicant lesion using shape, texture, and intensity. At that point, Support Vector Machine (SVM) classifier is applied with different cross validations on the features set to compare the precision of proposed framework. The proposed method is validated on three benchmark datasets such as Harvard, RIDER and Local. The method achieved average 97.1% accuracy, 0.98 area under curve, 91.9% sensitivity and 98.0% specificity. It can be used to identify the tumor more accurately in less processing time as compared to existing methods.