CAS-GAN for Contrast-free Angiography Synthesis
It addresses a critical safety issue for patients undergoing medical imaging by reducing reliance on harmful contrast agents, though it appears incremental as it builds on existing GAN and disentanglement methods.
This paper tackles the problem of health risks from iodinated contrast agents in interventional procedures by proposing CAS-GAN, a GAN framework that synthesizes X-ray angiographies without contrast, achieving a FID of 5.87 and MMD of 0.016 on the XCAD dataset.
Iodinated contrast agents are widely utilized in numerous interventional procedures, yet posing substantial health risks to patients. This paper presents CAS-GAN, a novel GAN framework that serves as a "virtual contrast agent" to synthesize X-ray angiographies via disentanglement representation learning and vessel semantic guidance, thereby reducing the reliance on iodinated contrast agents during interventional procedures. Specifically, our approach disentangles X-ray angiographies into background and vessel components, leveraging medical prior knowledge. A specialized predictor then learns to map the interrelationships between these components. Additionally, a vessel semantic-guided generator and a corresponding loss function are introduced to enhance the visual fidelity of generated images. Experimental results on the XCAD dataset demonstrate the state-of-the-art performance of our CAS-GAN, achieving a FID of 5.87 and a MMD of 0.016. These promising results highlight CAS-GAN's potential for clinical applications.