CycleGuardian: A Framework for Automatic RespiratorySound classification Based on Improved Deep clustering and Contrastive Learning

2025-02-02

AI Summary

To address challenges including few-shot learning, high noise levels, difficulty in fine-grained anomaly discrimination, and resource constraints for edge deployment, this paper proposes a lightweight intelligent auscultation method for respiratory sound analysis under resource-limited conditions. The method innovatively integrates hybrid spectrogram generation with a grouping strategy, designs a similarity-constrained deep clustering module to enhance abnormal feature representation, and introduces group-wise mixed contrastive learning to improve fine-grained discriminability. It employs multi-objective joint optimization and a lightweight CNN architecture. Evaluated on the ICBHI2017 dataset, the model achieves specificity of 82.06%, sensitivity of 44.47%, and a composite Score of 63.26%—surpassing the state-of-the-art by approximately 7 percentage points—while maintaining only 38 MB of parameters and enabling successful deployment on Android mobile devices.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Auscultation plays a pivotal role in early respiratory and pulmonary disease diagnosis. Despite the emergence of deep learning-based methods for automatic respiratory sound classification post-Covid-19, limited datasets impede performance enhancement. Distinguishing between normal and abnormal respiratory sounds poses challenges due to the coexistence of normal respiratory components and noise components in both types. Moreover, different abnormal respiratory sounds exhibit similar anomalous features, hindering their differentiation. Besides, existing state-of-the-art models suffer from excessive parameter size, impeding deployment on resource-constrained mobile platforms. To address these issues, we design a lightweight network CycleGuardian and propose a framework based on an improved deep clustering and contrastive learning. We first generate a hybrid spectrogram for feature diversity and grouping spectrograms to facilitating intermittent abnormal sound capture.Then, CycleGuardian integrates a deep clustering module with a similarity-constrained clustering component to improve the ability to capture abnormal features and a contrastive learning module with group mixing for enhanced abnormal feature discernment. Multi-objective optimization enhances overall performance during training. In experiments we use the ICBHI2017 dataset, following the official split method and without any pre-trained weights, our method achieves Sp: 82.06 $%$, Se: 44.47$%$, and Score: 63.26$%$ with a network model size of 38M, comparing to the current model, our method leads by nearly 7$%$, achieving the current best performances. Additionally, we deploy the network on Android devices, showcasing a comprehensive intelligent respiratory sound auscultation system.

Problem

• Automatic Respiratory Sound Recognition
• Limited Data Challenge
• Model Deployment on Small Devices

Innovation

• CycleGuardian
• Contrastive Learning
• Smart Segmentation and Clustering