Medical Image Generation and Augmentation with Generative Adversarial Networks (GAN’s)

Abstract

This project explores the use of Generative Adversarial
Networks (GANs) to address this challenge by generating realistic synthetic medical images that can augment existing datasets. GANs, composed of a generator and a discriminator in a competitive framework, can create high-fidelity images that resemble real medical images, such as X-rays, MRIs, and CT scans.
The proposed project focuses on developing a GAN model optimized for medical image synthesis and augmentation, improving its capacity to generate domain-specific features crucial for accurate diagnosis. Additionally, the model's synthetic data will be evaluated for quality and utility by training classification and segmentation models to assess the efficacy of GAN-augmented datasets in improving diagnostic performance. This project aims to facilitate a robust framework that mitigates data scarcity, supports enhanced model training, and ultimately contributes to the broader field of AI-driven healthcare solutions.
Single image super-resolution (SISR) has played an important role in the field of image processing. Recent generative adversarial networks (GANs) can achieve excellent results on low-resolution images with small samples. However, there are little literatures summarizing different GANs in SISR. In this paper, we conduct a comparative study of GANs from different perspectives. We first take a look at developments of GANs. Second, we present popular architectures for GANs in big and small samples for image applications. Single image super- resolution (SISR) has played an important role in the field of image processing. Recent generative adversarial networks (GANs) can achieve excellent results on low- resolution images with small samples.