{"title":"咳嗽声分类对早期哮喘筛查的探讨。","authors":"Yanming Huo, Jiajing Ma, Huixian Liu, Luyuan Jia, Guo Zhang, Congkang Zhang, Xu Guo, Shen-Ao Hao, Yongdong Song, Haotian Sun","doi":"10.1007/s11882-025-01213-w","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose of review: </strong>This review aims to explore an effective and scalable approach for early asthma detection using cough sounds. The main objective is to evaluate whether a multi-model deep learning fusion framework can improve diagnostic accuracy and generalizability in real-world settings.</p><p><strong>Recent findings: </strong>Recent research in respiratory sound analysis has demonstrated the potential of deep learning models in detecting pulmonary diseases. However, most studies focus on single-network architectures and often overlook class imbalance and training stability, which can limit model performance in practical applications. This study presents an asthma detection model that integrates ResNet18, VGG16, and DenseNet121 through a fusion layer. SMOTE is used to address data imbalance, and a weighted cross-entropy loss enhances training robustness. Mixed precision training and StepLR scheduling further improve performance. The proposed model achieved 95.9% accuracy on the test set, demonstrating strong generalization and potential for real-time, non-invasive screening in clinical environments.</p>","PeriodicalId":55198,"journal":{"name":"Current Allergy and Asthma Reports","volume":"25 1","pages":"36"},"PeriodicalIF":4.6000,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Investigation into the Classification of Cough Sounds for Early Asthma Screening.\",\"authors\":\"Yanming Huo, Jiajing Ma, Huixian Liu, Luyuan Jia, Guo Zhang, Congkang Zhang, Xu Guo, Shen-Ao Hao, Yongdong Song, Haotian Sun\",\"doi\":\"10.1007/s11882-025-01213-w\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Purpose of review: </strong>This review aims to explore an effective and scalable approach for early asthma detection using cough sounds. The main objective is to evaluate whether a multi-model deep learning fusion framework can improve diagnostic accuracy and generalizability in real-world settings.</p><p><strong>Recent findings: </strong>Recent research in respiratory sound analysis has demonstrated the potential of deep learning models in detecting pulmonary diseases. However, most studies focus on single-network architectures and often overlook class imbalance and training stability, which can limit model performance in practical applications. This study presents an asthma detection model that integrates ResNet18, VGG16, and DenseNet121 through a fusion layer. SMOTE is used to address data imbalance, and a weighted cross-entropy loss enhances training robustness. Mixed precision training and StepLR scheduling further improve performance. The proposed model achieved 95.9% accuracy on the test set, demonstrating strong generalization and potential for real-time, non-invasive screening in clinical environments.</p>\",\"PeriodicalId\":55198,\"journal\":{\"name\":\"Current Allergy and Asthma Reports\",\"volume\":\"25 1\",\"pages\":\"36\"},\"PeriodicalIF\":4.6000,\"publicationDate\":\"2025-08-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Current Allergy and Asthma Reports\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1007/s11882-025-01213-w\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ALLERGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current Allergy and Asthma Reports","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s11882-025-01213-w","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ALLERGY","Score":null,"Total":0}
Investigation into the Classification of Cough Sounds for Early Asthma Screening.
Purpose of review: This review aims to explore an effective and scalable approach for early asthma detection using cough sounds. The main objective is to evaluate whether a multi-model deep learning fusion framework can improve diagnostic accuracy and generalizability in real-world settings.
Recent findings: Recent research in respiratory sound analysis has demonstrated the potential of deep learning models in detecting pulmonary diseases. However, most studies focus on single-network architectures and often overlook class imbalance and training stability, which can limit model performance in practical applications. This study presents an asthma detection model that integrates ResNet18, VGG16, and DenseNet121 through a fusion layer. SMOTE is used to address data imbalance, and a weighted cross-entropy loss enhances training robustness. Mixed precision training and StepLR scheduling further improve performance. The proposed model achieved 95.9% accuracy on the test set, demonstrating strong generalization and potential for real-time, non-invasive screening in clinical environments.
期刊介绍:
The aim of Current Allergy and Asthma Reports is to systematically provide the views of highly selected experts on current advances in the fields of allergy and asthma and highlight the most important papers recently published. All reviews are intended to facilitate the understanding of new advances in science for better diagnosis, treatment, and prevention of allergy and asthma.
We accomplish this aim by appointing international experts in major subject areas across the discipline to review select topics emphasizing recent developments and highlighting important new papers and emerging concepts. We also provide commentaries from well-known figures in the field, and an Editorial Board of internationally diverse members suggests topics of special interest to their country/region and ensures that topics are current and include emerging research. Over a one- to two-year period, readers are updated on all the major advances in allergy and asthma.