CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning

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 abnor...

Full description

Saved in:
Bibliographic Details
Main Authors: Yun Chu, Qiuhao Wang, Enze Zhou, Ling Fu, Qian Liu, Gang Zheng
Format: Article
Language:English
Published: Springer 2025-03-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01800-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849392600030642176
author Yun Chu
Qiuhao Wang
Enze Zhou
Ling Fu
Qian Liu
Gang Zheng
author_facet Yun Chu
Qiuhao Wang
Enze Zhou
Ling Fu
Qian Liu
Gang Zheng
author_sort Yun Chu
collection DOAJ
description 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 facilitate 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 38 M. Compared 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.
format Article
id doaj-art-0eb523ca5c244a54bc7ee4aa887f9e9a
institution Kabale University
issn 2199-4536
2198-6053
language English
publishDate 2025-03-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-0eb523ca5c244a54bc7ee4aa887f9e9a2025-08-20T03:40:44ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-03-0111412010.1007/s40747-025-01800-4CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learningYun Chu0Qiuhao Wang1Enze Zhou2Ling Fu3Qian Liu4Gang Zheng5School of Information and Communication, Hainan UniversitySchool of BioMedical Engineering, Hainan UniversitySchool of Electronic Information and Communications, Huazhong University of Science and TechnologySchool of BioMedical Engineering, Hainan UniversitySchool of BioMedical Engineering, Hainan UniversitySchool of Electronic Information and Communications, Huazhong University of Science and TechnologyAbstract 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 facilitate 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 38 M. Compared 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.https://doi.org/10.1007/s40747-025-01800-4Intelligent auscultationDeep learningContrastive learningRespiratory sound
spellingShingle Yun Chu
Qiuhao Wang
Enze Zhou
Ling Fu
Qian Liu
Gang Zheng
CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
Complex & Intelligent Systems
Intelligent auscultation
Deep learning
Contrastive learning
Respiratory sound
title CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
title_full CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
title_fullStr CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
title_full_unstemmed CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
title_short CycleGuardian: a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
title_sort cycleguardian a framework for automatic respiratory sound classification based on improved deep clustering and contrastive learning
topic Intelligent auscultation
Deep learning
Contrastive learning
Respiratory sound
url https://doi.org/10.1007/s40747-025-01800-4
work_keys_str_mv AT yunchu cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning
AT qiuhaowang cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning
AT enzezhou cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning
AT lingfu cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning
AT qianliu cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning
AT gangzheng cycleguardianaframeworkforautomaticrespiratorysoundclassificationbasedonimproveddeepclusteringandcontrastivelearning