Super-resolution is a technique that enhances image quality by transforming lowresolution images into high-resolution representations. In medical imaging, image quality plays a crucial role in early diagnosis and clinical workflow efficiency. However, due to hardware limitations, low-dose acquisition requirements, and patient motion, medical images often suffer from significant information loss. This study presents a comprehensive review of image super-resolution methods based on Generative Adversarial Networks. Within the scope of this review, SR techniques are systematically categorized, and both Single-Image Super-Resolution and Multi-Image Super-Resolution approaches are analyzed in detail. The findings from the reviewed studies are summarized and presented in tabular form. Finally, the advantages, challenges, and limitations of GAN-based SR methods in clinical applications are discussed, providing insights for future research in medical image reconstruction and enhancement.
Goven et al. (Thu,) studied this question.