Semantic Image Synthesis for Abdominal CT
This work provides an incremental improvement for medical imaging researchers by enhancing data augmentation capabilities in abdominal CT synthesis.
The authors tackled the problem of generating abdominal CT images from semantic masks using conditional diffusion models, achieving better image quality compared to GAN-based approaches and finding that separate encoding of mask and input is more effective than concatenation.
As a new emerging and promising type of generative models, diffusion models have proven to outperform Generative Adversarial Networks (GANs) in multiple tasks, including image synthesis. In this work, we explore semantic image synthesis for abdominal CT using conditional diffusion models, which can be used for downstream applications such as data augmentation. We systematically evaluated the performance of three diffusion models, as well as to other state-of-the-art GAN-based approaches, and studied the different conditioning scenarios for the semantic mask. Experimental results demonstrated that diffusion models were able to synthesize abdominal CT images with better quality. Additionally, encoding the mask and the input separately is more effective than naïve concatenating.