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

TL;DR

This paper introduces PC-MCL, a novel multi-cycle learning approach for respiratory sound classification that corrects label bias, preserves information, and enhances generalization, achieving state-of-the-art results on the ICBHI benchmark.

Contribution

The paper proposes a new 3-label formulation, multi-cycle concatenation, and a patient-matching auxiliary task to improve respiratory sound classification and address label bias and overfitting issues.

Findings

Achieves 65.37% ICBHI Score, outperforming baselines.

Demonstrates the effectiveness of the 3-label formulation.

Shows all three components are essential through ablation studies.

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.