Automatic brain tumor segmentation based on deep learning methods: A review

Brain tumor segmentation (BTS) represents a challenging task in medical image processing. BTS aims to establish a precise delineation of brain tumor regions using masks that are accurately located. Manual segmentation has proven to be ineffective with a large dataset in terms of efficiency and time consumption. Recently, researchers have used deep learning techniques to segment the tumor from MRI images in a fully automatic way and without human intervention. In this paper, we seek to achieve two objectives. Firstly, we aim to introduce new deep learning based BTS methods that have been recently developed. Convolutional Neural Network (CNN) methods were used to automatically segment the brain tumor through the use of public and well-accepted dataset BRATS. Secondly providing a review for the segmentation of MRI brain tumors based on deep learning methods and discuss the final results.