提出C-Score评估医学影像模型解释一致性，发现模型稳定性问题。
Quantifying Explanation Consistency: The C-Score Metric for CAM-Based Explainability in Medical Image Classification
- 用强度强化的软交并比衡量同类病例解释重复性，无需人工标注。
- 在30轮训练中检测到三种一致性与准确率脱节现象，提前预警模型崩溃。
- 适合关注模型可解释性与临床部署安全性的研究人员使用。
类激活映射（CAM）方法广泛用于医学影像分类模型的可视化解释。然而，现有评估框架仅关注解释是否正确（通过与放射科医生标注的定位保真度衡量），而忽略了其一致性：即模型在不同患者间对同一病理是否采用一致的空间推理策略。本文提出C-Score（一致性评分），一种无需标注、基于置信度加权的度量指标，通过正确分类样本间的强度强调型软交并比来量化类别内解释的可复现性。我们在三个CNN架构（DenseNet201、InceptionV3、ResNet50V2）上，对六种CAM方法（GradCAM、GradCAM++、LayerCAM、EigenCAM、ScoreCAM、MS GradCAM++）在Kermany胸部X光数据集上进行了30轮训练周期的评估，覆盖迁移学习与微调阶段。发现了三种标准分类指标无法察觉的一致性-准确率分离机制：阈值导致的黄金列表坍缩、特定技术在峰值AUC时的归因坍缩，以及全局聚合中的类别级一致性掩盖。C-Score能提供模型不稳定性的早期预警信号。例如，ScoreCAM在ResNet50V2上的退化可在灾难性AUC崩溃前整整一个检查点被识别，从而给出基于解释质量而非预测排名的架构特定临床部署建议。
原文摘要 · Abstract (English)
Class Activation Mapping (CAM) methods are widely used to generate visual explanations for deep learning classifiers in medical imaging. However, existing evaluation frameworks assess whether explanations are correct, measured by localisation fidelity against radiologist annotations, rather than whether they are consistent: whether the model applies the same spatial reasoning strategy across different patients with the same pathology. We propose the C-Score (Consistency Score), a confidence-weighted, annotation-free metric that quantifies intra-class explanation reproducibility via intensity-emphasised pairwise soft IoU across correctly classified instances. We evaluate six CAM techniques: GradCAM, GradCAM++, LayerCAM, EigenCAM, ScoreCAM, and MS GradCAM++ across three CNN architectures (DenseNet201, InceptionV3, ResNet50V2) over thirty training epochs on the Kermany chest X-ray dataset, covering transfer learning and fine-tuning phases. We identify three distinct mechanisms of AUC-consistency dissociation, invisible to standard classification metrics: threshold-mediated gold list collapse, technique-specific attribution collapse at peak AUC, and class-level consistency masking in global aggregation. C-Score provides an early warning signal of impending model instability. ScoreCAM deterioration on ResNet50V2 is detectable one full checkpoint before catastrophic AUC collapse and yields architecture-specific clinical deployment recommendations grounded in explanation quality rather than predictive ranking alone.
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。