PulseAugur
EN
LIVE 09:13:16

LUCID framework uses Flow Matching for sparse-view CT reconstruction

Researchers have introduced LUCID, a novel framework for reconstructing high-quality computed tomography (CT) images from sparse-view data. This method utilizes Flow Matching for generative modeling, enabling it to adapt to varying levels of undersampling without requiring specific training for each setting. LUCID learns a continuous transport between a Gaussian distribution and high-quality CT images, and during inference, it incorporates the sparsity level to modulate the generative process, improving image quality and structural fidelity while reducing the risk of hallucinated structures. AI

RANK_REASON The cluster contains an academic paper detailing a new method for image reconstruction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

COVERAGE [1]

  1. arXiv cs.AI TIER_1 English(EN) · Jigang Duan, Jiayi Wang, Heran Wang, Ping Yang, Genwei Ma, Xing Zhao ·

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

    arXiv:2606.16212v1 Announce Type: cross 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 detail…