中文标题#
DINOv3 用于医学图像配准的测试时训练
英文标题#
DINOv3 with Test-Time Training for Medical Image Registration
中文摘要#
先验的医学图像配准方法,特别是基于学习的方法,通常需要大量训练数据,这限制了临床应用。 为了克服这一限制,我们提出了一种无需训练的流程,该流程依赖于冻结的 DINOv3 编码器和特征空间中的变形场测试时优化。 在两个代表性基准上,该方法准确且产生规则的变形。 在 Abdomen MR-CT 上,它达到了最佳平均 Dice 分数(DSC)0.790,同时具有最低的 95 百分位 Hausdorff 距离(HD95)4.9+-5.0 和 Log-Jacobian 的标准差(SDLogJ)0.08+-0.02。 在 ACDC 心脏 MRI 上,它将平均 DSC 提高到 0.769,并将 SDLogJ 降低到 0.11,HD95 降低到 4.8,相对于初始对齐有显著提升。 结果表明,在测试时在一个紧凑的基础特征空间中操作为临床配准提供了一个实用且通用的解决方案,而无需额外训练。
英文摘要#
Prior medical image registration approaches, particularly learning-based methods, often require large amounts of training data, which constrains clinical adoption. To overcome this limitation, we propose a training-free pipeline that relies on a frozen DINOv3 encoder and test-time optimization of the deformation field in feature space. Across two representative benchmarks, the method is accurate and yields regular deformations. On Abdomen MR-CT, it attained the best mean Dice score (DSC) of 0.790 together with the lowest 95th percentile Hausdorff Distance (HD95) of 4.9+-5.0 and the lowest standard deviation of Log-Jacobian (SDLogJ) of 0.08+-0.02. On ACDC cardiac MRI, it improves mean DSC to 0.769 and reduces SDLogJ to 0.11 and HD95 to 4.8, a marked gain over the initial alignment. The results indicate that operating in a compact foundation feature space at test time offers a practical and general solution for clinical registration without additional training.
文章页面#
PDF 获取#
抖音扫码查看更多精彩内容