CVMay 25
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, Nika Rasoolzadeh, Xin Wang, Yuan Gao, Muzhen He
arXiv:2605.2576746.2
Predicted impact top 72% in CV · last 90 daysOriginality Incremental advance
AI Analysis
For radiologists and breast cancer screening, this method improves the fidelity of synthetic contrast-enhanced MRI, potentially reducing contrast agent exposure.
SAFE-Diff introduces a diffusion model with scale-aware attention and feature-dispersive regularization to synthesize contrast-enhanced breast MRI, achieving a PSNR of 28.3 dB and SSIM of 0.91, outperforming prior methods by 2.1 dB.
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.