# Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions

arXiv: https://arxiv.org/abs/2605.29862 · HTML: https://mlsanity.com/paper/2605.29862/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 34 references that are in the map.
- First posted on arXiv 28 May 2026.

## Builds on (most cited first, 19 of 19)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [In Search of Lost Domain Generalization](https://mlsanity.com/paper/2007.01434/) (2020)
- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [AST: Audio Spectrogram Transformer](https://mlsanity.com/paper/2104.01778/) (2021)
- [PromptFL: Let Federated Participants Cooperatively Learn Prompts Instead of Models -- Federated Learning in Age of Foundation Model](https://mlsanity.com/paper/2208.11625/) (2022)
- [Causality-inspired Single-source Domain Generalization for Medical Image Segmentation](https://mlsanity.com/paper/2111.12525/) (2021)
- [Patch-Mix Contrastive Learning with Audio Spectrogram Transformer on Respiratory Sound Classification](https://mlsanity.com/paper/2305.14032/) (2023)
- [Stethoscope-guided Supervised Contrastive Learning for Cross-domain Adaptation on Respiratory Sound Classification](https://mlsanity.com/paper/2312.09603/) (2023)
- [RepAugment: Input-Agnostic Representation-Level Augmentation for Respiratory Sound Classification](https://mlsanity.com/paper/2405.02996/) (2024)
- [BTS: Bridging Text and Sound Modalities for Metadata-Aided Respiratory Sound Classification](https://mlsanity.com/paper/2406.06786/) (2024)
- [A Comprehensive Survey on Generative Diffusion Models for Structured Data](https://mlsanity.com/paper/2306.04139/) (2023)
- [Next Visit Diagnosis Prediction via Medical Code-Centric Multimodal Contrastive EHR Modelling with Hierarchical Regularisation](https://mlsanity.com/paper/2401.11648/) (2024)
- [Federated Out-of-Distribution Generalization: A Causal Augmentation View](https://mlsanity.com/paper/2504.19882/) (2025)
- [Empowering Multimodal Respiratory Sound Classification with Counterfactual Adversarial Debiasing for Out-of-Distribution Robustness](https://mlsanity.com/paper/2510.22263/) (2025)
- [Overcoming Uncertain Incompleteness for Robust Multimodal Sequential Diagnosis Prediction via Curriculum Data Erasing Guided Knowledge Distillation](https://mlsanity.com/paper/2407.19540/) (2024)
- [Lungmix: A Mixup-Based Strategy for Generalization in Respiratory Sound Classification](https://mlsanity.com/paper/2501.00064/) (2024)

## Built on by (most cited first, 1 of 1)

- [Quality Adaptive Angular Margin Learning for Respiratory Sound Classification](https://mlsanity.com/paper/2606.11915/) (2026)

## Abstract (arXiv)

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 in which clients hold recordings from different stethoscopes and the model is evaluated on an unseen device. Our empirical analysis shows that stethoscope-induced style and disease-relevant content are partially entangled, making deterministic style removal unreliable. In response, we propose BTS-CAFE, a framework combining (i) causality-inspired device-style interventions with constraints designed to limit content distortion, (ii) counterfactual metadata augmentation to relieve device and demographic shortcuts, and (iii) gradient alignment to promote device-invariant decision boundaries across clients. Built on BTS with CLAP, a multimodal language-audio pretraining model, BTS-CAFE improves the out-of-distribution ICBHI Score by 3.69 points on average over five held-out devices relative to its backbone, and outperforms conventional data augmentation and federated learning baselines in simulated device leave-out evaluations on the ICBHI and SPRSound datasets.