Wheezing and crackles in respiratory diseases
Pro Research Analysisby 
Searched over 200M research papers
Wheezing and Crackles: Key Respiratory Disease Indicators
Wheezing and crackles are abnormal lung sounds, known as adventitious sounds, that play a crucial role in diagnosing and monitoring respiratory diseases. These sounds are commonly detected through auscultation and are associated with various lung conditions, but can also be present in healthy individualsAviles-Solis2019Solis2018.
Prevalence and Clinical Associations of Wheezing and Crackles
Large population studies show that wheezes and crackles are common, with about 28% of adults exhibiting these sounds during lung auscultationAviles-Solis2019Solis2018. Wheezes are more often heard during expiration, while crackles are typically detected during inspiration. Age is the strongest predictor for both sounds, especially crackles, with prevalence increasing as people get olderAviles-Solis2019Solis2018. Other significant associations include current smoking, self-reported respiratory diseases (such as asthma), reduced lung function, and lower oxygen saturationAviles-Solis2019Solis2018. The presence of these sounds in multiple lung areas is linked to a higher risk of decreased lung function.
Diagnostic Value in Respiratory Diseases
Wheezes are often linked to asthma and chronic obstructive pulmonary diseases, while crackles are more associated with pneumonia and bronchitis. Both sounds are important for distinguishing between different respiratory conditions and for assessing disease severityChen2019Aviles-Solis2019Solis2018. In bronchiectasis, the number of crackles and wheezes can be reliably measured and used to monitor disease status over time.
Advances in Automated Detection and Classification
Recent advances in machine learning and deep learning have significantly improved the accuracy and consistency of detecting and classifying wheezes and crackles from lung sound recordingsAmose2023Seong2024Fernandes2022+3 MORE. Deep learning models, such as convolutional neural networks (CNNs) and sequence models like LSTM and GRU, have demonstrated high accuracy in distinguishing between normal, wheeze, and crackle sounds, often outperforming human clinicians, especially less experienced onesAmose2023Fernandes2022Chen2019+2 MORE. These automated systems can help reduce diagnostic errors and support rapid, objective assessment in clinical and telemedicine settingsAmose2023Chen2019Kim2021.
Machine Learning in Non-Auscultatory Diagnosis
Machine learning techniques can also analyze cough sounds to differentiate between wheezing and crackles, offering a non-auscultatory approach to diagnosis. When combined with clinical features, these models achieve high sensitivity and specificity, suggesting potential for smart applications usable by non-medical personnel.
Impact of Demographics and Equipment
The accuracy of automated classification can vary based on factors such as age, gender, body mass index, and the type of recording equipment usedFernandes2022Amose2024. Age- and gender-specific models tend to perform better than general models, highlighting the importance of personalized approaches in lung sound analysis.
Conclusion
Wheezing and crackles are prevalent and clinically significant indicators in respiratory diseases, closely associated with age, smoking, and lung function. Automated analysis using advanced machine learning and deep learning methods offers reliable, objective, and scalable tools for detecting and classifying these sounds, supporting improved diagnosis and monitoring of respiratory conditions in both clinical and remote settingsAmose2023Seong2024Fernandes2022+7 MORE.
Sources and full results
Most relevant research papers on this topic
Wheeze and Crackle Analysis Using Deep Learning
Deep learning techniques can improve accuracy and consistency in analyzing wheeze and crackle sounds in respiratory diseases, offering a promising alternative to human auscultation.
DOI