Cross Modality Image Translation In Medical Imaging Using Generative Frameworks
The research compared seven state-of-the-art generative models for cross-modality 3D medical image translation across eleven datasets and three anatomical regions. The study established a standardized framework to rigorously evaluate these models, finding that GANs generally outperformed diffusion models in quantitative metrics like PSNR and SSIM, though perceptual realism was assessed via physician testing.