Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916061090152448 |
|---|---|
| author | Koo, Heejoon Kim, Yoon Tae Toikkanen, Miika Kim, June-Woo |
| author_facet | Koo, Heejoon Kim, Yoon Tae Toikkanen, Miika Kim, June-Woo |
| contents | AI-driven respiratory sound classification (RSC) is promising for automated pulmonary disease detection, yet multi-site deployment is hindered by inter-stethoscope variability. We introduce a federated domain generalization (FedDG) formulation for RSC under stethoscope-induced device shifts, where clients use heterogeneous devices and the model is evaluated on unseen devices. Our empirical analysis shows that stethoscope-induced style and disease-specific content are tightly entangled, making deterministic style removal unreliable. In response, we propose a causality-inspired multimodal FedDG framework that combines: (i) a causality-inspired device style intervention network that performs content-preserving style perturbations, (ii) counterfactual text augmentation that neutralizes metadata shortcuts, and (iii) gradient alignment that facilitates device-invariant representations across clients. Built on a multimodal language-audio pretraining model, it outperforms conventional data augmentation and federated learning baselines in leave-one-device-out validation on ICBHI and SPRSound datasets. Code will be released upon publication. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_29862 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions Koo, Heejoon Kim, Yoon Tae Toikkanen, Miika Kim, June-Woo Audio and Speech Processing Artificial Intelligence Sound AI-driven respiratory sound classification (RSC) is promising for automated pulmonary disease detection, yet multi-site deployment is hindered by inter-stethoscope variability. We introduce a federated domain generalization (FedDG) formulation for RSC under stethoscope-induced device shifts, where clients use heterogeneous devices and the model is evaluated on unseen devices. Our empirical analysis shows that stethoscope-induced style and disease-specific content are tightly entangled, making deterministic style removal unreliable. In response, we propose a causality-inspired multimodal FedDG framework that combines: (i) a causality-inspired device style intervention network that performs content-preserving style perturbations, (ii) counterfactual text augmentation that neutralizes metadata shortcuts, and (iii) gradient alignment that facilitates device-invariant representations across clients. Built on a multimodal language-audio pretraining model, it outperforms conventional data augmentation and federated learning baselines in leave-one-device-out validation on ICBHI and SPRSound datasets. Code will be released upon publication. |
| title | Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions |
| topic | Audio and Speech Processing Artificial Intelligence Sound |
| url | https://arxiv.org/abs/2605.29862 |