Glioma Detection and Segmentation Using Deep Learning Architectures

Abstract

Glioma is the primary type of brain tumors which occurred in both human brain regions and spinal cord. It can be cured if it is detected on time by the current scanning methodologies. In this paper, effective Computer Assisted Artificial Methods (CAAM) using deep learning architectures are proposed to differentiate the pixels belonging to Glioma tumors and the pixels belong to non-Glioma tumors. The LeNET and AlexNET deep learning structures along with morphological segmentation procedures are applied on the brain images to detect and locate the tumor pixels in Glioma images. The experimental results are carried out on BRTAS 2019 and BRATS 2020 dataset brain Magnetic Resonance Imaging (MRI). The experimental results of this work are extensively analyzed and compared with similar existing studies