Generative Modeling in Sinogram Domain for Sparse-view CT Reconstruction

TL;DR

This paper introduces an unsupervised score-based generative model operating in the sinogram domain to improve sparse-view CT reconstruction, reducing the need for paired training data and achieving high-quality images.

Contribution

The authors propose a novel unsupervised generative approach in the sinogram domain that does not require paired data for training, enabling flexible sparse-view CT reconstruction.

Findings

Achieved comparable or superior results to supervised methods

Effectively reduces artifacts in sparse-view CT images

Operates without retraining for different view configurations

Abstract

The radiation dose in computed tomography (CT) examinations is harmful for patients but can be significantly reduced by intuitively decreasing the number of projection views. Reducing projection views usually leads to severe aliasing artifacts in reconstructed images. Previous deep learning (DL) techniques with sparse-view data require sparse-view/full-view CT image pairs to train the network with supervised manners. When the number of projection view changes, the DL network should be retrained with updated sparse-view/full-view CT image pairs. To relieve this limitation, we present a fully unsupervised score-based generative model in sinogram domain for sparse-view CT reconstruction. Specifically, we first train a score-based generative model on full-view sinogram data and use multi-channel strategy to form highdimensional tensor as the network input to capture their prior distribution. Then, at the inference stage, the stochastic differential equation (SDE) solver and data-consistency step were performed iteratively to achieve fullview projection. Filtered back-projection (FBP) algorithm was used to achieve the final image reconstruction. Qualitative and quantitative studies were implemented to evaluate the presented method with several CT data. Experimental results demonstrated that our method achieved comparable or better performance than the supervised learning counterparts.