发表机构
German Cancer Research Center (DKFZ); Heidelberg University; HIDSS4Health – Helmholtz Information and Data Science School for Health; Helmholtz Imaging（德国癌症研究中心; 海德堡大学; HIDSS4Health – 亥姆霍兹健康信息与数据科学学校; 亥姆霍兹成像中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对DCE-MRI峰值增强合成任务，提出Anguinus Sculpturae组合式流水线，利用分割指导的单步差异预测，在MAMA-MIA Duke子集上取得最佳FRD和Dice，兼顾全局保真与病灶结构。
AI 中文摘要
动态对比增强乳腺MRI（DCE-MRI）富含解剖和灌注信息，但其对钆基对比剂的依赖引发了安全性担忧并增加了成本。虚拟对比增强，即从预对比图像合成后对比图像，是一种有前景的替代方案。我们针对MAMA-SYNTH挑战任务，即预测峰值增强乳腺MRI。我们没有采用扩散或流匹配的完整机制，而是观察到在矫正的直线路径下，生成过程简化为单一的差异预测：合成峰值图像等于预对比图像加上预测的增强图，通过一次前向传播即可获得。在此基础上，我们构建了Anguinus Sculpturae，一个组合式流水线，其中nnU-Net对病灶、前景和乳腺区域的分割指导两个生成器——一个优化全局保真度，另一个通过冻结分割器路由的非对称Tversky项优化病灶结构——并通过高斯加权混合进行区域级合成。在MAMA-MIA的留出Duke子集上，我们的模型在所有评估变体中取得了最佳的FRD和Dice分数，表明单步差异预测结合分割指导足以恢复全局保真度和病灶结构。代码可在该https URL获取。
英文摘要
Dynamic contrast-enhanced breast MRI (DCE-MRI) is rich in anatomical and perfusion information, but its reliance on gadolinium-based contrast agents raises safety concerns and adds cost. Virtual contrast enhancement, synthesizing post-contrast from pre-contrast images, is a promising alternative. We address the MAMA-SYNTH challenge task of predicting peak-enhancement breast MRI. Rather than adopting the full machinery of diffusion or flow matching, we observe that under a rectified, straight-line path the generative process collapses to a single difference prediction: the synthetic peak image is the pre-contrast image plus a predicted enhancement map, recovered in one forward pass. Around this we build Anguinus Sculpturae, a compositional pipeline in which nnU-Net segmentations of lesion, foreground and breast region guide two generators - one optimized for global fidelity, one for lesion structure through an asymmetric Tversky term routed via a frozen segmenter - composited region-wise with Gaussian-weighted blending. On the held-out Duke subset of MAMA-MIA our model achieves the best FRD and Dice among all evaluated variants, showing that single-step difference prediction with segmentation guidance suffices to recover both global fidelity and lesion structure. Code is available at https://github.com/MIC-DKFZ/AnguinusSculpturae.
CommentsAccepted as an oral at the MAMA-SYNTH 2026 challenge / Deep-BreAth 2026 Workshop, MICCAI 2026. 12 pages, 3 figures, 1 table
