用小波流匹配实现超快多模态MRI生成，一步到位。
WFM: 3D Wavelet Flow Matching for Ultrafast Multi-Modal MRI Synthesis
- 从先验均值出发，直接学习小波空间的映射路径。
- 仅需1-2步积分，速度比扩散模型快250-1000倍。
- 单模型支持四类脑肿瘤影像，适合临床实时应用。
扩散模型在多模态MRI合成中表现优异，但计算开销大（数百次采样步骤，每模态需独立模型），限制了临床部署。我们发现其低效源于起点不合理：扩散从纯噪声开始，忽略了已有序列中的结构信息。为此提出WFM（小波流匹配），学习从条件模态的小波空间均值到目标分布的直接映射。因源与目标共享解剖结构、仅对比度不同，该方法可在1-2步积分内实现高精度合成。一个8200万参数的类别条件模型可生成全部四类BraTS模态（T1, T1c, T2, FLAIR），替代总计3.26亿参数的四个独立扩散模型。在BraTS 2024上，WFM达到26.8 dB PSNR和0.94 SSIM，性能仅落后扩散基线1-2 dB，而推理时间仅为0.16-0.64秒/体积（相比基线160秒）。这一速度-质量平衡使实时MRI合成成为临床可行方案。代码已开源。
原文摘要 · Abstract (English)
Diffusion models have achieved remarkable quality in multi-modal MRI synthesis, but their computational cost (hundreds of sampling steps and separate models per modality) limits clinical deployment. We observe that this inefficiency stems from an unnecessary starting point: diffusion begins from pure noise, discarding the structural information already present in available MRI sequences. We propose WFM (Wavelet Flow Matching), which instead learns a direct flow from an informed prior, the mean of conditioning modalities in wavelet space, to the target distribution. Because the source and target share underlying anatomy and differ primarily in contrast, this formulation enables accurate synthesis in just 1-2 integration steps. A single 82M-parameter model with class conditioning synthesizes all four BraTS modalities (T1, T1c, T2, FLAIR), replacing four separate diffusion models totaling 326M parameters. On BraTS 2024, WFM achieves 26.8 dB PSNR and 0.94 SSIM, within 1-2 dB of diffusion baselines, while running 250-1000x faster (0.16-0.64s vs. 160s per volume). This speed-quality trade-off makes real-time MRI synthesis practical for clinical workflows. Code is available at https://github.com/yalcintur/WFM.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。