Zhifeng Wang, Renjiao Yi, Xin Wen, Chenyang Zhu, Kai Xu, Kunlun He
The Angio-Diff model uses a self-supervised diffusion approach to transform non-angiographic X-rays into high-quality angiographic images, addressing data shortages and improving vascular image synthesis accuracy.
Vascular diseases are serious health concerns, and X-ray angiography is a key diagnostic tool, but it involves high radiation exposure. To reduce this risk, researchers are developing methods to convert safer, non-angiographic X-rays into detailed angiographic images using advanced AI models. The Angio-Diff model uses a new approach to create realistic images of blood vessels from non-angiographic X-rays, overcoming challenges like data scarcity and ensuring the accurate depiction of complex vascular structures. This method could significantly improve the safety and accessibility of vascular imaging.
