🤖 AI Summary
To address severe image quality degradation in ultra-sparse-angle CT reconstruction, this paper proposes MSDiff, a multi-scale diffusion model. Methodologically, MSDiff introduces a novel multi-scale score-based diffusion framework incorporating an isometric projection masking mechanism to explicitly encode intrinsic correlations among projection measurements; it further integrates adaptive noise scheduling and cross-scale feature fusion to jointly preserve global anatomical structures and recover local fine details. Crucially, MSDiff is the first to unify full-sampling prior knowledge with selectively sparse-view information within the diffusion process, thereby overcoming the performance limitations of conventional score-based generative models under extremely low-angle regimes (e.g., 16 or 32 views). Extensive experiments on multiple public CT datasets demonstrate significant improvements in reconstruction fidelity—quantified by PSNR and SSIM—and strong generalizability across diverse anatomies and sparsity levels. This work establishes a new paradigm for low-dose CT imaging.
📝 Abstract
Computed Tomography (CT) technology reduces radiation haz-ards to the human body through sparse sampling, but fewer sampling angles pose challenges for image reconstruction. Score-based generative models are widely used in sparse-view CT re-construction, performance diminishes significantly with a sharp reduction in projection angles. Therefore, we propose an ultra-sparse view CT reconstruction method utilizing multi-scale dif-fusion models (MSDiff), designed to concentrate on the global distribution of information and facilitate the reconstruction of sparse views with local image characteristics. Specifically, the proposed model ingeniously integrates information from both comprehensive sampling and selectively sparse sampling tech-niques. Through precise adjustments in diffusion model, it is capable of extracting diverse noise distribution, furthering the understanding of the overall structure of images, and aiding the fully sampled model in recovering image information more effec-tively. By leveraging the inherent correlations within the projec-tion data, we have designed an equidistant mask, enabling the model to focus its attention more effectively. Experimental re-sults demonstrated that the multi-scale model approach signifi-cantly improved the quality of image reconstruction under ultra-sparse angles, with good generalization across various datasets.