Brain tumor segmentation using support vector machine (SVM)

Abstract

Brain tumor segmentation is one of the most difficult procedures in the medical field, it involves extracting tumor patches from images. One non-invasive method that has emerged as a leading non-ionizing radiation brain tumor diagnosis tool is magnetic resonance imaging (MRI). Gliomas are the most frequent and aggressive type of brain tumor, with a very short survival time in their maximum degree. Manual segmentation takes a lot of time in clinical practice, and the operator's expertise has a big impact on how well they execute. This project proposes fully automatic segmentation of brain tumor using Support Vector Machine(SVM). Further, it uses brain image from BRATS 2020 database .BRATS 2020 focuses on the segmentation of fundamentally heterogeneous (in appearance, form, and histology) brain tumors, specifically gliomas, using multi-institutional pre-operative MRI data. Hence, this project segments brain tumor into two classes like tumor region and non-tumor region.Brain tumor segmentation needs to separate healthy tissues from tumor region to distinguish between normal and abnormal pixels. This is an essential step in diagnosis and treatment planning, both of which need to take place quickly in case of a malignancy in order to maximize the likelihood of successful treatment.