Limited Angle

Limited-angle tomography addresses the challenge of reconstructing high-quality images from incomplete X-ray projection data, a problem arising in various applications like medical imaging and materials science where full angular scans are impractical or impossible. Current research focuses on leveraging deep learning, particularly diffusion models and neural networks, often incorporating multi-scale approaches or low-resolution priors to improve reconstruction accuracy and efficiency. These advancements aim to mitigate artifacts and improve the quality of images obtained from limited data, ultimately enhancing the capabilities and expanding the applications of tomography techniques.