[论文解读] Interpretable Aneurysm Classification via 3D Concept Bottleneck Models: Integrating Morphological and Hemodynamic Clinical Features

Toqa Khaled, Ahmad Al-Kabbany|arXiv (Cornell University)|Mar 8, 2026

Intracranial Aneurysms: Treatment and Complications被引用 0

一句话总结

该论文提出一个用于颅内动脉瘤分类的三维软概念瓶颈模型（Soft-CBM），使用CTA，整合三维形态学与CFD派生的血流动力学概念，在可解释的临床瓶颈下实现高准确性。

ABSTRACT

We are concerned with the challenge of reliably classifying and assessing intracranial aneurysms using deep learning without compromising clinical transparency. While traditional black-box models achieve high predictive accuracy, their lack of inherent interpretability remains a significant barrier to clinical adoption and regulatory approval. Explainability is paramount in medical modeling to ensure that AI-driven diagnoses align with established neurosurgical principles. Unlike traditional eXplainable AI (XAI) methods -- such as saliency maps, which often provide post-hoc, non-causal visual correlations -- Concept Bottleneck Models (CBMs) offer a robust alternative by constraining the model's internal logic to human-understandable clinical indices. In this article, we propose an end-to-end 3D Concept Bottleneck framework that maps high-dimensional neuroimaging features to a discrete set of morphological and hemodynamic concepts for aneurysm identification. We implemented this pipeline using a pre-trained 3D ResNet-34 backbone and a 3D DenseNet-121 to extract features from CTA volumes, which were subsequently processed through a soft bottleneck layer representing human-interpretable clinical concepts. The model was optimized using a joint-loss function to balance diagnostic focal loss and concept mean squared error (MSE), validated via stratified five-fold cross-validation. Our results demonstrate a peak task classification accuracy of 93.33% +/- 4.5% for the ResNet-34 architecture and 91.43% +/- 5.8% for the DenseNet-121 model. Furthermore, the implementation of 8-pass Test-Time Augmentation (TTA) yielded a robust mean accuracy of 88.31%, ensuring diagnostic stability during inference. By maintaining an accuracy-generalization gap of less than 0.04, this framework proves that high predictive performance can be achieved without sacrificing interpretability.

研究动机与目标

• 在颅内动脉瘤分类中，将高预测性能与临床可解释性相结合。
• 将3D成像与血流动力学特征映射到人类可解释的临床概念集合。
• 通过分阶段训练和多级数据增强，减少数据泄漏并提升泛化能力。

提出的方法

• 使用3D骨干网络（在MedicalNet上预训练的ResNet-34与从头训练的DenseNet-121）从CTA提取体积特征。
• 用一个同一嵌入的最终层替代并添加一个软瓶颈，预测26个临床概念。
• 将潜在视觉嵌入与预测概念拼接用于最终诊断（z ⊕ c）。
• 用联合损失优化：Ltotal = β Ltask + α Lconcept，其中β=1.0，α=0.01。
• 使用分阶段微调：先冻结编码器再解冻，并使用5折分层交叉验证。
• 采用多级3D增强及8次测试时增强（TTA）以提高鲁棒性。

实验结果

研究问题

• RQ1 3D软概念瓶颈模型在动脉瘤分类上是否能够实现高准确性，同时提供可解释的临床推理？
• RQ2 使用预训练与从头训练的3D骨干在该CBM框架中对准确性与泛化性的影响？
• RQ3 多级增强与TTA是否在不牺牲可解释性的前提下提高推理稳定性？
• RQ4 将概念约束为26个临床有意义指标对性能与泄漏缓解有何影响？

主要发现

• 任务峰值准确率：ResNet-34与合并策略达到93.33% ± 4.5%。
• 任务峰值准确率：DenseNet-121从头训练达到91.43% ± 5.8%。
• 8次TTA实现平均准确率88.31%，推理鲁棒性良好。
• 跨试验的准确性-泛化差距≤0.04，表明过拟合较低。
• ResNet-34合并配置的ROC AUC均值0.960 ± 0.032；TTA为0.899，DenseNet-121为0.899。
• 标准推理中ResNet-34合并的灵敏度最高约97.8%，TTA提升特异性。