Brain Tumor Segmentation of Brain MRI Using Watershed Algorithm

Abstract

One of the life-threatening disease affecting the brain is the brain cancer. Manual segmentation is more time consuming and being susceptible to human bias or mistakes. Detection of the tumor at an early stage becomes necessary in order to save lives. Brain tumor can be detected with the help of MR images. In this paper, we have proposed a method for brain tumor segmentation using deep learning. Here to extract and segment the tumor we used threshold segmentation and watershed segmentation. After the segmentation, morphological operations like image opening, image closing, dilation and erosion were applied to remove the unwanted parts. As a result some useful information in simpler form will be presented in front of the users, especially for the medical staff treating the patient.