发布日期: 2025-09-21

更新日期: 2025-09-22

文章字数: 2k

阅读时长: 8 分

阅读次数:

⚠️ 以下所有内容总结都来自于大语言模型的能力，如有错误，仅供参考，谨慎使用
🔴 请注意：千万不要用于严肃的学术场景，只能用于论文阅读前的初筛！
💗 如果您觉得我们的项目对您有帮助 ChatPaperFree ，还请您给我们一些鼓励！⭐️ HuggingFace免费体验

2025-09-21 更新

DICE: Diffusion Consensus Equilibrium for Sparse-view CT Reconstruction

Authors:Leon Suarez-Rodriguez, Roman Jacome, Romario Gualdron-Hurtado, Ana Mantilla-Dulcey, Henry Arguello

Sparse-view computed tomography (CT) reconstruction is fundamentally challenging due to undersampling, leading to an ill-posed inverse problem. Traditional iterative methods incorporate handcrafted or learned priors to regularize the solution but struggle to capture the complex structures present in medical images. In contrast, diffusion models (DMs) have recently emerged as powerful generative priors that can accurately model complex image distributions. In this work, we introduce Diffusion Consensus Equilibrium (DICE), a framework that integrates a two-agent consensus equilibrium into the sampling process of a DM. DICE alternates between: (i) a data-consistency agent, implemented through a proximal operator enforcing measurement consistency, and (ii) a prior agent, realized by a DM performing a clean image estimation at each sampling step. By balancing these two complementary agents iteratively, DICE effectively combines strong generative prior capabilities with measurement consistency. Experimental results show that DICE significantly outperforms state-of-the-art baselines in reconstructing high-quality CT images under uniform and non-uniform sparse-view settings of 15, 30, and 60 views (out of a total of 180), demonstrating both its effectiveness and robustness.

稀疏视图计算机断层扫描（CT）重建技术因其欠采样问题而面临根本性挑战，导致逆向问题表述不清晰。传统的迭代方法虽然通过引入手工制作或学习的先验来约束解的形式，但仍难以捕获医学影像中复杂的结构特征。与此不同，扩散模型（DMs）作为一种强大的生成先验模型，能够精确地模拟复杂的图像分布，最近开始崭露头角。在这项工作中，我们引入了扩散共识均衡（DICE）框架，该框架将双智能体共识均衡融入扩散模型的采样过程中。DICE交替进行以下操作：（i）数据一致性智能体，通过近端算子强制实施测量一致性；（ii）先验智能体，通过扩散模型在每个采样步骤中进行干净图像估计。通过平衡这两个互补智能体的迭代交互，DICE有效地结合了强大的生成先验能力和测量一致性。实验结果表明，在15、30和60个视图的均匀和非均匀稀疏视图设置下，DICE在重建高质量CT图像方面显著优于当前先进的基线方法，这证明了其有效性和稳健性。

论文及项目相关链接

PDF 8 pages, 4 figures, confenrence

Summary

基于扩散模型，我们提出了一个新的CT重建框架DICE。DICE结合了两种关键思想：通过近端操作器强制测量一致性的数据一致性代理和通过扩散模型进行清洁图像估计的先验代理。通过迭代平衡这两个互补的代理，DICE有效地结合了强大的先验能力和测量一致性。在均匀和非均匀稀疏视图条件下，DICE显著优于现有技术基线，在重建高质量CT图像方面表现出卓越的效果和稳健性。

Key Takeaways

DICE是一个新的CT重建框架，基于扩散模型（DMs）。
DICE结合了数据一致性代理和先验代理两种思想。
数据一致性代理通过近端操作器强制测量一致性。
先验代理通过扩散模型进行清洁图像估计。
DICE框架通过迭代平衡这两个互补的代理，有效地结合了强大的先验能力和测量一致性。
在均匀和非均匀稀疏视图条件下，DICE显著优于现有技术基线。

Cool Papers

点此查看论文截图

First Detection of X-ray Polarization and Its Short-term Increase Pre- and Post-Eclipse in HMXB 4U 1700-377

Authors:Kaito Ninoyu, Keisuke Uchiyama, Shinya Yamada, Ryota Hayakawa, Shunji Kitamoto, Nao Kominato, Takayoshi Kohmura, Misaki Mizumoto, Yuusuke Uchida, Toru Tamagawa, Ryota Tomaru, Seoru Ito

We report the first statistically significant detection of X-ray polarization from the high-mass X-ray binary (HMXB) 4U 1700-377, observed using the Imaging X-ray Polarimetry Explorer (IXPE). A polarization degree exceeding 10% was detected above 5 keV, placing it among the highest polarization observed in HMXBs to date. The observation was conducted over a full orbital period of the binary system, during which several sporadic and instantaneous flares were detected. We identify a clear correlation between the polarization degree and orbital phase, with the highest polarization occurring just before and after the eclipse, reaching over 20% for a few tens of ks. These results suggest that the scattering medium responsible for the observed polarization is spatially localized between the compact object and the O-type companion star, likely created by large-scale inhomogeneities in the stellar wind and its interaction with the compact star’s emission. We also explore the roles of disk winds and orbital reflection in the observed polarization variability. While both mechanisms contribute to the polarization, the substantial increase in polarization before and after the eclipse cannot be fully explained by these models alone, suggesting that the involvement of additional factors. The properties of the X-ray polarization observed by IXPE provide new insights into the accretion processes, X-ray emission, and wind structure in 4U 1700-377, advancing our understanding of their complex environments and the nature of the compact objects within.

我们报告了使用成像X射线偏振仪（IXPE）观察到的来自高质量X射线双星（HMXB）4U 1700-377的X射线偏振的首次统计学显著检测结果。在大于5 keV的能段，检测到超过10%的偏振度，这是迄今为止在HMXB中观察到的最高偏振度之一。观测是在双星系统的整个轨道周期中进行的，在此期间检测到了几次零星和瞬间的耀斑。我们发现偏振度与轨道相位之间存在明显的相关性，偏振度最高的时刻出现在凌星之前和之后，持续数十千秒，达到超过20%。这些结果表明，造成所观察到的偏振的散射介质位于致密天体与O型伴星之间，可能是由恒星风的大规模不均匀性及其与致密恒星发射的相互作用所造成的。我们还探讨了圆盘风和轨道反射对观察到的偏振变化的作用。虽然这两种机制都有助于偏振，但在凌星前后出现的偏振度大幅上升无法仅通过这些模型来解释，这表明还有其他因素的参与。IXPE观察到的X射线偏振特性为4U 1700-377的吸积过程、X射线发射和风结构提供了新见解，推动了我们对其复杂环境和内部致密天体的本质的理解。

论文及项目相关链接

PDF

Summary

本文报道了使用成像X射线偏振探测器（IXPE）首次从高质量X射线双星（HMXB）4U 1700-377检测到具有统计学意义的X射线偏振，其偏振度超过10%。观测发现偏振度与轨道相位存在明确关联，最高偏振出现在蚀前和蚀后，持续数十千秒。这些结果暗示散射介质可能位于致密天体与O型伴星之间，可能是由恒星风的大规模不均匀性及其与致密天体发射的相互作用所形成。此外，本文还探讨了圆盘风和轨道反射对偏振变化的影响，但仍有待进一步解释偏振度的显著增加。

Key Takeaways