Improving Limited Angle CT Reconstruction with a Robust GAN Prior
This work addresses the problem of limited angle CT reconstruction for medical imaging, offering a scanner-agnostic method that is incremental in applying a robust GAN prior.
The paper tackles limited angle CT reconstruction by using a robust GAN prior to clean noisy, artifact-laden reconstructions, achieving significant improvements in reconstruction quality through a corruption mimicking technique.
Limited angle CT reconstruction is an under-determined linear inverse problem that requires appropriate regularization techniques to be solved. In this work we study how pre-trained generative adversarial networks (GANs) can be used to clean noisy, highly artifact laden reconstructions from conventional techniques, by effectively projecting onto the inferred image manifold. In particular, we use a robust version of the popularly used GAN prior for inverse problems, based on a recent technique called corruption mimicking, that significantly improves the reconstruction quality. The proposed approach operates in the image space directly, as a result of which it does not need to be trained or require access to the measurement model, is scanner agnostic, and can work over a wide range of sensing scenarios.