发表机构
Image Sciences Institute; University Medical Center Utrecht（影像科学研究所; 乌得勒支大学医学中心）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出先验引导的混合概念瓶颈模型，在0-20%的低概念标注场景下，可提升癌症影像诊断的概念AUC并保持诊断性能，同时降低标注负担。
AI 中文摘要
概念瓶颈模型（CBMs）可通过放射学概念表达预测结果，从而提升癌症影像诊断预测的透明度。然而，这类模型对实例级概念标注的依赖限制了其实际应用。本文提出一种先验引导的混合CBM，该模型整合了有限的概念标注、未标注患者的类条件概念分布匹配，以及概念到诊断头的先验初始化。我们在CBIS-DDSM乳腺肿块与钙化灶、LIDC-IDRI肺结节数据集上，针对0-100%的概念标注比例对该方法进行评估。在临床相关的0-20%标注范围内，混合CBM的平均概念AUC始终优于匹配的标准CBM，同时保持与黑箱模型相近的诊断性能。具体而言，在10%标注比例下，肿块的概念AUC从0.619提升至0.741，钙化灶从0.650提升至0.787，肺结节从0.597提升至0.642。消融实验表明，先验初始化是提升概念检测的主要组成部分，其作用可能在于稳定概念到诊断头。此外，零样本视觉语言模型（VLMs）对于可靠的细粒度肿瘤级概念预测仍显不足。这些发现表明，结构化先验可大幅降低可解释癌症影像模型的标注负担。
英文摘要
Concept bottleneck models (CBMs) can improve the transparency of cancer image diagnostic prediction by expressing predictions through radiological concepts. However, their dependence on instance-level concept annotations limits practical applicability. We propose a prior-guided hybrid CBM that integrates limited concept annotations, class-conditional concept distribution matching on unannotated patients, and prior initialization of the concept-to-diagnosis head. We evaluate the method on CBIS-DDSM mammographic masses and calcifications and LIDC-IDRI pulmonary nodules across 0-100% concept annotation. In the clinically relevant 0-20% annotation regime, the hybrid CBM consistently improves mean concept AUC over a matched standard CBM, while maintaining diagnostic performance close to black-box models. At 10% annotation specifically, concept AUC increases from 0.619 to 0.741 for masses, from 0.650 to 0.787 for calcifications, and from 0.597 to 0.642 for pulmonary nodules. Ablation experiments identify prior initialization as the main component contributing to improved concept detection, likely by stabilizing the concept-to-diagnosis head. Zero-shot VLMs remain insufficient for reliable fine-grained tumor-level concept prediction. These findings suggest that structured priors can substantially reduce the annotation burden of interpretable cancer imaging models.
CommentsAccepted at the iMIMIC Workshop at MICCAI 2026
