Please use this identifier to cite or link to this item:
Title: Brain Tumor Segmentation by Generative Adversarial Network (GAN)
Authors: Guluzade, Jale
Keywords: Brain tumors
Segmentation methods
Magnetic Resonance Imaging (MRI) images
Generative Adversarial Networks (GAN)
Low-grade gliomas (LGG)
High-grade gliomas (HGG)
Image resolutions
Issue Date: 2021
Abstract: The concept of a brain tumor is one of the most significant health issues in terms of both economic and social stability. This disease is extensive growth of abnormal cells in the brain and any growth inside can lead to any serious problem. The cost of a patient’s life is a primary concern, so multiple monitoring and treatment systems are still improving to build up the long-term life expectancy of the better life of those patients who have severe brain tumor problems. However, there exists a lack of data available associated with medical diagnosis and images in which intensive diagnostic analytics (DA) techniques are demanded today. In these cases, accurate performance improvement is a major factor of positive enhancement in treatment and diagnostics by the fact that a lack of medical images has constant distribution compared with real image distributions. Therefore, deep learning of structural variability of brain tumors substantially offers contrast-enhanced images to eliminate attainable data gaps and lacks in image distribution.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
Brain Tumor Segmentation by Generative Adversarial Network (GAN).pdf1.86 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.