提出可解释的超声心动图域偏移量化方法，预测分割性能下降。
Domain Shift in Echocardiography: Interpretable Quantification and Prediction of Cross-Dataset Left Ventricular Segmentation
- 用手工特征、变分自编码器和分割特征评估跨数据集迁移退化。
- 几何预处理显著改善较差迁移案例，亮度对比非主要偏移源。
- 不同表示下用特定统计距离预测性能下降，支持无掩码风险监控。
跨数据集泛化仍是超声心动图左心室分割临床部署的主要障碍，但其来源极少被拆解。本文在六个超声心动图数据集上，通过手工超声描述符、变分自编码器（VAE）潜在特征及分割衍生特征，检验了迁移退化是否可在部署前被估计。几何感知预处理显著改善多个表现不佳的迁移案例，表明大部分看似域偏移实则源于视野与构图不一致，而非固有声学差异。强度z标准化对数据集可分性影响小于0.005，说明亮度与对比度并非主导偏移轴。对保留数据源-目标对的绝对Dice分数下降，预测的决定系数R²达0.612，平均绝对误差MAE为0.082，斯皮尔曼等级相关系数rho为0.681。不含左心室与扇形特征的变体仍保持约70%解释力，支持无需掩码的迁移风险监测。最有效的差异度量取决于表示方式：在z标准化手工特征中，条件均值差（CMD）表现最强，绝对相关系数r≈0.86，R²≈0.70；在VAE空间中，对数瓦瑟斯坦距离最强，r≈-0.90，R²≈0.81；在左心室分割潜在特征中，对数最大均值差异（log-MMD）最强，r≈-0.92，R²≈0.84。表面厂商效应大多由数据集混淆所致。因此，超声心动图域偏移具有结构且可度量，其对分割的影响可通过几何感知预处理部分缓解，并借助特定表示的迁移风险估计提前预测。
原文摘要 · Abstract (English)
Cross-dataset generalisation remains a major barrier to clinical deployment of echocardiographic left ventricular segmentation, yet the sources of this shift are rarely disentangled. We examined whether transfer degradation could be estimated before deployment using handcrafted ultrasound descriptors, VAE latent features, and segmentation-derived latent features across six echocardiographic datasets. Geometry-aware preprocessing substantially improved several poor transfer cases, suggesting that much of the apparent domain shift reflects field-of-view and framing inconsistencies rather than intrinsic acoustic differences alone. Intensity z-normalisation changed dataset separability by less than 0.005, indicating that brightness and contrast are not the dominant shift axis. Absolute Dice drop on held-out source-target pairs was predicted with an R-squared value of 0.612, an MAE of 0.082, and a Spearman rho of 0.681. The variant without LV and fan-shaped features retained approximately 70% of this explanatory power, supporting mask-free transfer-risk monitoring. The most informative discrepancy measure depended on the representation, with CMD strongest in z-normalised handcrafted features, with an absolute r of approximately 0.86 and an R-squared value of approximately 0.70; log-Wasserstein strongest in VAE space, with an r of approximately -0.90 and an R-squared value of approximately 0.81; and log-MMD strongest in LV-segmentation latent features, with an r of approximately -0.92 and an R-squared value of approximately 0.84. Apparent vendor effects were largely dataset-confounded. Echocardiographic domain shift is therefore structured and measurable, and its impact on segmentation can be partly reduced through geometry-aware preprocessing and anticipated using representation-specific transfer-risk estimation.
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。