Glioblastoma brain tumor segmentation and survival prediction using 3D U Net

Abstract

Convolutional networks are the most widely used method for egmenting three dimensional (3D) medical images. Deep learning algorithms have recently attained human-level performance in a number of significant application tasks, including lung cancer volumetry and delineation ,preparing for radiation therapy However, cutting-edge topologies like UNet and Deep Medic are computationally intensive and necessitate workstations with graphics processing units for quick inference.