Physics-informed DeepCT: Sinogram Wavelet Decomposition Meets Masked Diffusion

TL;DR

This paper introduces SWARM, a novel sinogram wavelet decomposition and mask diffusion model that enhances sparse-view CT reconstruction by expanding training data variability and improving detail capture.

Contribution

The paper presents a new method combining sinogram wavelet decomposition with random masking and diffusion strategies to improve generalization and detail preservation in SVCT reconstruction.

Findings

SWARM outperforms existing methods in quantitative metrics.

Enhanced feature representation improves detail reconstruction.

Random mask strategy broadens training data diversity.

Abstract

Diffusion model shows remarkable potential on sparse-view computed tomography (SVCT) reconstruction. However, when a network is trained on a limited sample space, its generalization capability may be constrained, which degrades performance on unfamiliar data. For image generation tasks, this can lead to issues such as blurry details and inconsistencies between regions. To alleviate this problem, we propose a Sinogram-based Wavelet random decomposition And Random mask diffusion Model (SWARM) for SVCT reconstruction. Specifically, introducing a random mask strategy in the sinogram effectively expands the limited training sample space. This enables the model to learn a broader range of data distributions, enhancing its understanding and generalization of data uncertainty. In addition, applying a random training strategy to the high-frequency components of the sinogram wavelet enhances feature representation and improves the ability to capture details in different frequency bands, thereby improving performance and robustness. Two-stage iterative reconstruction method is adopted to ensure the global consistency of the reconstructed image while refining its details. Experimental results demonstrate that SWARM outperforms competing approaches in both quantitative and qualitative performance across various datasets.