无需重训练，让预训练医学图像配准模型自动生成误差图。
Uncertainty Estimation for Pretrained Medical Image Registration Models via Transformation Equivariance
- 利用配准变换的等变性，在推理时估算不确定性。
- 在3个模型、4类器官上，不确定性图与配准误差高度相关。
- 适合临床和大规模研究，提升预训练模型的安全性。
准确的图像配准在诸多医学影像应用中至关重要，但大多数深度配准网络无法提供其预测不可靠的提示。现有不确定性估计方法如贝叶斯方法、集成或MC-dropout通常需修改架构或重新训练，难以应用于预训练模型。本文提出一种推理时、模型无关的不确定性估计框架，可直接作用于任意预训练配准网络。该方法基于图像配准的变换等变性：输入的空间扰动下，解剖结构映射应保持一致。在三个预训练配准模型和四个解剖结构上的实验表明，生成的不确定性图与配准误差显著相关，能有效识别配准不稳定的区域。该框架使预训练配准模型在测试时具备风险感知能力，推动医学图像配准向安全的临床及大规模研究部署迈进。
原文摘要 · Abstract (English)
Accurate image registration is essential in many medical imaging applications, yet most deep registration networks provide little indication of when or where their predictions are unreliable. Existing uncertainty estimation approaches, such as Bayesian methods, ensembles, or MC-dropout, typically require architectural modifications or retraining, precluding their applicability to pretrained registration models. We propose an inference-time, model-agnostic uncertainty estimation framework that applies directly to any pretrained registration network. Our approach is grounded in the transformation equivariance property of image registration, which states that the underlying anatomical mapping should remain consistent under spatial perturbations of the input. Experiments across three pretrained registration models and four anatomical structures show that the resulting uncertainty maps consistently correlate with registration error and highlight unreliably aligned regions. This framework turns pretrained registration networks into risk-aware tools at test time, moving medical image registration closer to safe clinical and large-scale research deployment.
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。