用因果方法解决医疗影像跨医院泛化难题
CIV-DG: Conditional Instrumental Variables for Domain Generalization in Medical Imaging
- 基于条件工具变量，分离疾病特征与设备伪影
- 在多个医学影像数据集上显著超越现有基线
- 适合关注医疗AI泛化能力的研究者
医疗AI的跨机构泛化能力受选择偏差的根本性制约，这种结构性偏差表现为患者特征（如年龄、病情严重程度）非随机地决定医院归属。传统域泛化方法主要针对图像分布变化，无法解决由此产生的站点特异性变异与诊断标签之间的虚假相关。为此，我们提出CIV-DG，一种利用条件工具变量的因果框架，可解耦病理语义与扫描仪引入的伪影。该方法放宽了标准工具变量对随机分配的严格假设，适用于由患者特征内生驱动的临床场景。我们通过深度广义矩估计（DeepGMM）架构实现该理论，采用条件判别器最小化矩条件违反，并在人口统计亚组内强制工具变量与误差正交。在Camelyon17基准和大规模胸部X光数据集上的大量实验表明，CIV-DG显著优于主流基线，验证了条件因果机制在应对结构性混杂方面的有效性。
原文摘要 · Abstract (English)
Cross-site generalizability in medical AI is fundamentally compromised by selection bias, a structural mechanism where patient demographics (e.g., age, severity) non-randomly dictate hospital assignment. Conventional Domain Generalization (DG) paradigms, which predominantly target image-level distribution shifts, fail to address the resulting spurious correlations between site-specific variations and diagnostic labels. To surmount this identifiability barrier, we propose CIV-DG, a causal framework that leverages Conditional Instrumental Variables to disentangle pathological semantics from scanner-induced artifacts. By relaxing the strict random assignment assumption of standard IV methods, CIV-DG accommodates complex clinical scenarios where hospital selection is endogenously driven by patient demographics. We instantiate this theory via a Deep Generalized Method of Moments (DeepGMM) architecture, employing a conditional critic to minimize moment violations and enforce instrument-error orthogonality within demographic strata. Extensive experiments on the Camelyon17 benchmark and large-scale Chest X-Ray datasets demonstrate that CIV-DG significantly outperforms leading baselines, validating the efficacy of conditional causal mechanisms in resolving structural confounding for robust medical AI.
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。