SynthRAR: Ring Artifacts Reduction in CT with Unrolled Network and Synthetic Data Training

TL;DR

SynthRAR introduces an unrolled neural network approach that leverages synthetic data to effectively reduce ring artifacts in CT images, considering both sinogram and image domain correlations, without needing extensive real-world datasets.

Contribution

The paper proposes a novel unrolled network that models the CT forward operation and uses synthetic data to train for ring artifact reduction, addressing data collection costs and domain correlation.

Findings

Outperforms existing methods across various geometries and regions

Effectively leverages synthetic data for training without clinical data

Consistently improves artifact correction in diverse scenarios

Abstract

Defective and inconsistent responses in CT detectors can cause ring and streak artifacts in the reconstructed images, making them unusable for clinical purposes. In recent years, several ring artifact reduction solutions have been proposed in the image domain or in the sinogram domain using supervised deep learning methods. However, these methods require dedicated datasets for training, leading to a high data collection cost. Furthermore, existing approaches focus exclusively on either image-space or sinogram-space correction, neglecting the intrinsic correlations from the forward operation of the CT geometry. Based on the theoretical analysis of non-ideal CT detector responses, the RAR problem is reformulated as an inverse problem by using an unrolled network, which considers non-ideal response together with linear forward-projection with CT geometry. Additionally, the intrinsic correlations of ring artifacts between the sinogram and image domains are leveraged through synthetic data derived from natural images, enabling the trained model to correct artifacts without requiring real-world clinical data. Extensive evaluations on diverse scanning geometries and anatomical regions demonstrate that the model trained on synthetic data consistently outperforms existing state-of-the-art methods.