发表机构
Maastricht University; GROW – Research Institute for Oncology and Reproduction（马斯特里赫特大学; GROW肿瘤与生殖研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出潜在桥匹配框架，在MAMA-SYNTH挑战中从乳腺DCE-MRI造影前图像合成峰值增强图像，肿瘤条件变体性能优于造影前条件及LDM基线，为虚拟造影增强提供新方案。
AI 中文摘要
动态对比增强磁共振成像（DCE-MRI）是乳腺癌成像的核心手段，但钆剂给药会增加扫描负担，因此催生了包括合成造影生成在内的低造影替代方案。我们在MAMA-SYNTH挑战场景下提出了潜在桥匹配（LBM）框架，用于从造影前图像合成峰值增强乳腺DCE-MRI。与传统潜在扩散模型（LDM）从高斯噪声开始不同，所提模型学习造影前与峰值增强VAE潜变量配对间的条件桥。潜在UNet从中间桥状态预测到峰值增强潜变量的剩余修正，实现迭代细化，同时将轨迹锚定到患者特定解剖结构。我们在91例DUKE验证病例上评估了两种LBM条件变体，对于肿瘤条件变体，采用肿瘤掩码作为条件输入。与造影前条件相比，肿瘤条件提升了性能：均方误差（MSE）从1.023降至0.940，Frechet距离（FRD）从7.523降至4.716，肿瘤结构相似性指数（SSIM）从0.355升至0.429。肿瘤条件LBM在该验证队列上也优于所评估的LDM基线。这些结果表明，潜在桥匹配是一种有前景的造影前锚定虚拟造影增强方案，但仍需进一步工作验证泛化性并消除推理时对真实肿瘤掩码的依赖。
英文摘要
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is central to breast cancer imaging, but gadolinium administration increases scan burden and motivates contrast-reduced alternatives, including synthetic contrast generation. We propose a latent bridge matching (LBM) framework for synthesizing peak-enhanced breast DCE-MRI from pre-contrast images in the MAMA-SYNTH challenge setting. Instead of starting from Gaussian noise as in conventional latent diffusion models (LDMs), the proposed model learns a conditional bridge between paired pre-contrast and peak-enhanced VAE latents. A latent UNet predicts the remaining correction from intermediate bridge states to the peak-enhanced latent, enabling iterative refinement while keeping the trajectory anchored to patient-specific anatomy. We evaluated two LBM conditioning variants on 91 DUKE validation cases. For the tumor-conditioned variant, tumor masks were used as conditioning inputs. Tumor-conditioning improved performance compared with pre-contrast conditioning, reducing MSE from 1.023 to 0.940 and FRD from 7.523 to 4.716, while increasing tumor SSIM from 0.355 to 0.429. The tumor-conditioned LBM also outperformed the evaluated LDM baseline on this validation cohort. These results suggest that latent bridge matching is a promising pre-contrast-anchored formulation for virtual contrast enhancement, while further work is needed to validate generalization and remove dependence on ground-truth tumor masks at inference.
