SAFE-Diff: Scale-Aware Attention and Feature-Dispersive Diffusion with Uncertainty Estimation for Contrast-Enhanced Breast MRI Synthesis

Tianyu Zhang; Xinglong Liang; Jarek van Dijk; Luyi Han; Chunyao Lu; Antonio Portaluri; Xinghe Xie; Yaofei Duan

cs.CVMay 25, 2026

SAFE-Diff: Scale-Aware Attention and Feature-Dispersive Diffusion with Uncertainty Estimation for Contrast-Enhanced Breast MRI Synthesis

Authors: Tianyu Zhang, Xinglong Liang, Jarek van Dijk, Luyi Han, Chunyao Lu, Antonio Portaluri, Xinghe Xie, Yaofei Duan, +8 more

Organizations: Department of Medical Imaging, Radboud University Medical Center, Geert Grooteplein 10, 6525 GA, Nijmegen, The Netherlands · Department of Radiology, Netherlands Cancer Institute, Plesmanlaan 121, 1066May CX, Amsterdam, The Netherlands · Maastro Clinic, Dr. Tanslaan 12, 6229 ET, Maastricht, The Netherlands · Faculty of Applied Science, Macao Polytechnic University, 999078, Macao, China · Department of Radiation Oncology, Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands

Abstract

Synthesizing high fidelity contrast enhanced MRI is clinically valuable for safer and more efficient breast cancer screening, yet remains challenging due to complex lesion textures and heterogeneous enhancement patterns.

Explore similar work

MIRAGE: Multi-scale Lesion-Informed Representation with Auxiliary Guidance for MRI Contrast Enhancement
Jul 21, 2026Andrea Borghesi, Xin Wang, Jonas Teuwen +1Adaptive Histogram Equalization Lesion
Segmentation-Assisted Brain MRI Synthesis with Cross-Image Multi-Contrast Feature Memory Bank Retrieval Augmentation
Jun 7, 2026Wenwei Huang, Jia Wei, Jianlong ZhouMagnetic Resonance Imaging Medical Image Generation
Dense Temporal Contrast Synthesis via Conditioned Latent Transport
Jul 31, 2026Smriti Joshi, Apostolia Tsirikoglou, Daniel M. Lang +15Magnetic Resonance Imaging Medical Image Generation

Jul 21, 2026 · eess.IVJ/K move · Enter open · S save

MIRAGE: Multi-scale Lesion-Informed Representation with Auxiliary Guidance for MRI Contrast Enhancement

Andrea Borghesi, Xin Wang, Jonas Teuwen, George Yiasemis

Netherlands Cancer Institute

Inferring contrast enhancement from one pre-contrast breast MRI slice is underdetermined: post-contrast appearance contains physiological information that is not uniquely encoded in baseline anatomy. Optimizing only paired pixel fidelity can suppress uncertain lesion enhancement, whereas adversarial or stochastic generative objectives can favor realistic post-contrast appearance without guaranteeing patient-specific lesion fidelity. We introduce MIRAGE, a residual 2D U-Net that combines global reconstruction and perceptual losses with three forms of lesion-aware supervision available only during training: an asymmetric penalty for missed tumor enhancement, multi-scale auxiliary tumor segmentation, and guidance through a frozen post-contrast tumor segmentation nnU-Net. We evaluate the method on 301 cases from the multi-centre MAMA-SYNTH data using eight complementary image-, region-, radiomics-, and segmentation-based metrics. MIRAGE ranks first on six metrics and markedly improves downstream lesion localization over tuned pix2pix, conditional diffusion, and latent bridge-matching baselines. The generative alternatives retain advantages in LPIPS or contrast classification, revealing a clear fidelity-utility trade-off. Leave-one-in and leave-one-out ablations show that the losses are partly redundant for lesion localization but exert distinct effects on appearance, radiomics, and boundary accuracy. These results support task-aware synthesis while also showing that its apparent optimality is conditional on the downstream models and metrics used to define utility.

Adaptive Histogram Equalization Lesion Magnetic Resonance Imaging U-Net Brain Tumor Segmentation Multiscale

Open paper Ask about it PDF

Abstract

Explore similar work

MIRAGE: Multi-scale Lesion-Informed Representation with Auxiliary Guidance for MRI Contrast Enhancement

Segmentation-Assisted Brain MRI Synthesis with Cross-Image Multi-Contrast Feature Memory Bank Retrieval Augmentation

Dense Temporal Contrast Synthesis via Conditioned Latent Transport