PC-MCL: Patient-Consistent Multi-Cycle Learning with multi-label bias correction for respiratory sound classification

2026-01-23

AI Summary

This work addresses patient-specific overfitting in respiratory sound classification and the loss of normal signals and label distribution bias caused by multi-cycle concatenation under conventional binary labeling. To mitigate these issues, the authors propose a patient-consistent multi-cycle learning framework that employs a three-label formulation—normal, crackles, and wheezes—to better calibrate label distributions. The framework further incorporates a patient-matching auxiliary task as a multi-task regularization mechanism, effectively preserving normal components in mixed samples and enhancing feature robustness. Evaluated on the ICBHI 2017 benchmark, the model achieves an ICBHI Score of 65.37%, outperforming existing methods. Ablation studies confirm the necessity and synergistic contribution of each component in the proposed framework.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Multi-modal Vision

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Automated respiratory sound classification supports the diagnosis of pulmonary diseases. However, many deep models still rely on cycle-level analysis and suffer from patient-specific overfitting. We propose PC-MCL (Patient-Consistent Multi-Cycle Learning) to address these limitations by utilizing three key components: multi-cycle concatenation, a 3-label formulation, and a patient-matching auxiliary task. Our work resolves a multi-label distributional bias in respiratory sound classification, a critical issue inherent to applying multi-cycle concatenation with the conventional 2-label formulation (crackle, wheeze). This bias manifests as a systematic loss of normal signal information when normal and abnormal cycles are combined. Our proposed 3-label formulation (normal, crackle, wheeze) corrects this by preserving information from all constituent cycles in mixed samples. Furthermore, the patient-matching auxiliary task acts as a multi-task regularizer, encouraging the model to learn more robust features and improving generalization. On the ICBHI 2017 benchmark, PC-MCL achieves an ICBHI Score of 65.37%, outperforming existing baselines. Ablation studies confirm that all three components are essential, working synergistically to improve the detection of abnormal respiratory events.

Problem

• respiratory sound classification
• multi-label bias
• patient-specific overfitting
• normal signal loss
• multi-cycle analysis

Innovation

• multi-cycle learning
• multi-label bias correction
• patient-consistent representation
• respiratory sound classification
• auxiliary patient-matching task