Skip to main content
QUICK REVIEW

[论文解读] Likelihood-Separable Diffusion Inference for Multi-Image MRI Super-Resolution

Samuel W. Remedios, Zhangxing Bian|arXiv (Cornell University)|Jan 20, 2026
Advanced Neuroimaging Techniques and Applications被引用 0
一句话总结

论文通过证明独立测量的似然可分离性,将基于扩散的逆问题求解器推广到多图像 MRI 超分辨率(MISR),实现无需联合算子或再训练的 MISR,并引入带噪声权重的 MISR 变体,取得最先进结果。

ABSTRACT

Diffusion models are the current state-of-the-art for solving inverse problems in imaging. Their impressive generative capability allows them to approximate sampling from a prior distribution, which alongside a known likelihood function permits posterior sampling without retraining the model. While recent methods have made strides in advancing the accuracy of posterior sampling, the majority focuses on single-image inverse problems. However, for modalities such as magnetic resonance imaging (MRI), it is common to acquire multiple complementary measurements, each low-resolution along a different axis. In this work, we generalize common diffusion-based inverse single-image problem solvers for multi-image super-resolution (MISR) MRI. We show that the DPS likelihood correction allows an exactly-separable gradient decomposition across independently acquired measurements, enabling MISR without constructing a joint operator, modifying the diffusion model, or increasing network function evaluations. We derive MISR versions of DPS, DMAP, DPPS, and diffusion-based PnP/ADMM, and demonstrate substantial gains over SISR across $4 imes/8 imes/16 imes$ anisotropic degradations. Our results achieve state-of-the-art super-resolution of anisotropic MRI volumes and, critically, enable reconstruction of near-isotropic anatomy from routine 2D multi-slice acquisitions, which are otherwise highly degraded in orthogonal views.

研究动机与目标

  • 在 MRI 场景中动机化 MISR:多幅各向异性低分辨率体积可为单一高分辨率体积提供信息。
  • 形式化一个可分离的似然框架,使 MISR 在没有联合前向算子或不改变扩散模型的情况下进行修正。
  • 推导 DPS、DMAP、DPPS 的 MISR 版本,以及基于扩散的 PnP/ADMM。
  • 引入每个测量的噪声加权以融合异质获取。
  • 在 4x/8x/16x 垂直/平面的降解下展示改进的 MISR 性能,并实现从常规二维获取中接近各向同性的重建。

提出的方法

  • 证明多个 LR 测量的联合对数似然在观测之间可分离,从而实现对每个测量的独立修正。
  • 推导可在独立的 A_i 运算符与 y_i 测量下运行的 MISR 版本的 DPS、DMAP、DPPS,以及基于扩散的 PnP/ADMM。
  • 引入逆方差权重 w_i,用于对不同噪声水平和分辨率的测量进行梯度加权融合。
  • 使用带 DDIM 基采样器的 3D 体积扩散模型,以及用于从 x_t 估计 x_0 的流估计网络。
  • 在保持扩散模型不变的同时,对每个测量单独用数据一致性梯度更新 x_t,然后将它们相加。
  • 展示可接受的计算效率(在 RTX 6000 ADA 上 64 NFEs 下每个体积 ≤ 60 秒)。
Figure 2 : Qualitative results for a representative subject from the AIBL dataset. Row-wise labels designate LR inputs as “Measurements” and SR estimations named by method. Supercolumns group scale factors together. Within supercolumns, each column corresponds to SISR using only the axial acquisitio
Figure 2 : Qualitative results for a representative subject from the AIBL dataset. Row-wise labels designate LR inputs as “Measurements” and SR estimations named by method. Supercolumns group scale factors together. Within supercolumns, each column corresponds to SISR using only the axial acquisitio

实验结果

研究问题

  • RQ1在不构建联合前向算子或不重新训练扩散模型的情况下,是否可以在扩散式逆问题求解器中实现 MISR?
  • RQ2似然可分离性是否使 MISR 能进行每个测量的数据一致性修正?
  • RQ3每个测量的噪声权重对各向异性尺度因子下的 MISR 性能有何影响?
  • RQ4哪种 MISR 扩散式方法(DPS/DMAP/DPPS/PnP 变体)在 PSNR、SSIM、FID 的折衷中对各向同性 MRI 提供最佳平衡?
  • RQ5常规二维多切片获取是否能够有效实现接近各向同性的 MISR 扩散重建?

主要发现

  • 具有似然可分离性的 MISR 扩散可从独立的每测量算子获得修正,避免联合算子或模型再训练。
  • MISR 变体(DPS、DMAP、DPPS 和基于扩散的 PnP/ADMM)将扩散先验扩展到多图像 MISR。
  • 逆方差噪声加权在噪声增加时显著提升 MISR 性能。
  • 在 4x/8x/16x 的穿透降级下,MISR 方法显著优于单图像超分(PSNR 提升约 1–3 dB)。
  • DMAP 在报告的实验中在失真指标(PSNR/SSIM)和 FID 上始终优于其他方法。
  • MISR 使从常规二维多切片获取中实现近似各向同性解剖结构成为可能,潜在降低对长时间三维扫描的需求。
Figure 3 : Sagittal slices from four representative subjects are shown for $8\times$ scale factor LR inputs. Each displayed image is labeled for its contents. LR: low-resolution; AX: axial acquisition; COR: coronal acquisition; HR: ground-truth high-resolution; SISR: single-image super-resolution, u
Figure 3 : Sagittal slices from four representative subjects are shown for $8\times$ scale factor LR inputs. Each displayed image is labeled for its contents. LR: low-resolution; AX: axial acquisition; COR: coronal acquisition; HR: ground-truth high-resolution; SISR: single-image super-resolution, u

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。