用反事实对抗方法让呼吸音分类更抗偏见，提升跨医院适用性。
Empowering Multimodal Respiratory Sound Classification with Counterfactual Adversarial Debiasing for Out-of-Distribution Robustness
- 基于因果图的反事实去偏，消除年龄性别等元数据干扰
- 对抗训练学习对元数据不敏感的特征，提升泛化能力
- 适合医疗多模态研究者，尤其关注模型鲁棒性的场景
多模态呼吸音分类通过融合生物声学信号与患者元数据，有望实现肺部疾病的早期检测。然而，现有方法易受年龄、性别或采集设备等属性引起的虚假相关性影响，导致在不同临床机构间分布偏移时性能下降。为此，本文提出一种反事实对抗去偏框架：首先，采用基于因果图的反事实去偏方法抑制元数据中的非因果依赖；其次，引入对抗去偏机制学习对元数据不敏感的表征，减少特定元数据偏差；第三，设计反事实元数据增强策略进一步缓解虚假相关性，强化元数据不变表征。实验表明，该方法在分布内及分布外条件下均显著优于强基线。代码已开源于 https://github.com/RSC-Toolkit/BTS-CARD。
原文摘要 · Abstract (English)
Multimodal respiratory sound classification offers promise for early pulmonary disease detection by integrating bioacoustic signals with patient metadata. Nevertheless, current approaches remain vulnerable to spurious correlations from attributes such as age, sex, or acquisition device, which hinder their generalization, especially under distribution shifts across clinical sites. To this end, we propose a counterfactual adversarial debiasing framework. First, we employ a causal graph-based counterfactual debiasing methodology to suppress non-causal dependencies from patient metadata. Second, we introduce adversarial debiasing to learn metadata-insensitive representations and reduce metadata-specific biases. Third, we design counterfactual metadata augmentation to mitigate spurious correlations further and strengthen metadata-invariant representations. By doing so, our method consistently outperforms strong baselines in evaluations under both in-distribution and distribution shifts. Code is available at https://github.com/RSC-Toolkit/BTS-CARD.
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。