Lumber Spine MRI Super-Resolution Using SRGAN’s

Abstract

Medical imaging is essential to modern healthcare because it provides accurate visual information for diagnosis and treatment planning. However, low-resolution images are often created due to inherent limitations of imaging technology, which impairs the ability of healthcare professionals to see small details. To overcome this difficulty, super-resolution - a method of improving image resolution - has become popular. This project investigates the use of Super-Resolution Generative Adversarial Networks (SRGAN) for medical image enhancement. In practice, adversarial training SRGANs, a method based on deep learning, produces high-resolution images of their low-resolution counterparts. In several fields, including medical imaging, the model's ability to learn complex mappings between low and high-resolution image spaces has produced impressive results. In this work, we specifically discuss how SRGANs can improve the resolution of medical images to help doctors diagnose patients with greater accuracy and detail. In this paper, we have discussed about our system that increases the resolution of low-resolution MRI imaging of Lumber Spine with high accuracy.