Glioblastoma detection using vgg16, inceptionnet, alexnet, resnet, and their comparative analysis

Abstract

In this study we propose a deep learning based model for the detection of glioblastoma(an aggressive type of cancer that can occur in the brain or spinal cord) based on their origin in brain . The deep learnig based model will be obtained from the comparative analysis of the mdoels resnet,vgg16 , inception- net,alexnet .We also attempt to overcome the shortcoming of the above paper( Identi cation of Glioma from MR Images Using Convolutional Neural Network ) [1] which is some astrocytomas and oligodendrocytomas are misidenti ed as GBM and do a comparative analysis between the implemented model.We conducted our experiment using the datasets TCGA and LGG1p19q Depletion from Cancerar- chieve,Medpix,BraTS2020 consisting of samples over more than 12,000 patients. The experiments using Glioma images from the Brats2020 shows that we obtain 86% average classi cation accuracy for the network .