LUCID: Learned Undersampling-Adaptive Consistency-Guided Inference with Deterministic Flow Matching for Sparse-View CT Reconstruction

2026-06-15

AI Summary

Sparse-view CT reconstruction suffers from streaking artifacts and anatomical distortions due to angular undersampling, and existing methods often either rely on specific sampling patterns or generate inconsistent hallucinated structures under severe undersampling. This work proposes LUCID, a novel framework that, for the first time, integrates deterministic flow matching with a sampling-adaptive mechanism. Trained solely on high-quality CT images, LUCID achieves stable reconstructions across diverse sparse-view settings through sparsely weighted initial state construction, sparsity-modulated flow matching updates, and projection-domain data consistency constraints. The method substantially improves image quality and structural fidelity while effectively suppressing hallucination artifacts and reducing the risk of anatomical inconsistency.

Abstract

Sparse-view CT reduces radiation dose and scanning time by acquiring fewer projection views, but angular undersampling makes reconstruction severely ill-posed, causing streak artifacts, structural blurring, and loss of fine details. Existing supervised methods are often tied to specific sampling settings, whereas generative methods may introduce anatomically inconsistent hallucination-like structures under severe undersampling. We propose Lucid, a sparsity-adaptive, consistency-guided reconstruction framework based on a Flow Matching generative prior for sparse-view CT. Lucid is trained only on high-quality CT images to learn a continuous transport between a Gaussian distribution and the high-quality CT image distribution, independent of view sampling. During inference, the sampling sparsity level is explicitly incorporated to adapt the generative trajectory of a single pretrained model. Specifically, Lucid constructs a degradation-matched initial state by sparsity-weighted fusion of the sparse-view FBP image and Gaussian noise, performs sparsity-modulated Flow Matching updates, and applies projection-domain data-consistency correction after each prior update. Experiments under multiple sparse-view settings show that Lucid achieves stable reconstruction performance across different sampling densities, improves image quality and structural fidelity, and reduces the risk of hallucination-like structures in generative sparse-view CT reconstruction.

Problem

• sparse-view CT
• angular undersampling
• ill-posed reconstruction
• hallucination artifacts
• streak artifacts

Innovation

• Flow Matching
• sparsity-adaptive
• data consistency
• generative prior
• sparse-view CT