Audio-Based Screening of Respiratory Diseases Using Machine Learning: A Methodological Framework Evaluated on a Clinically Validated COVID-19 Cough Dataset
Machine Learning and Knowledge Extraction · 20.03.2026 · 8 (3) · 80
DOI: 10.3390/make8030080 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machine Learning and Knowledge Extraction, online erschienen am 20.03.2026. Verfasst von Arley Magnolia Aquino-García und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 108 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Council on Humanities, Sciences and Technologies.
Abstract (Originalsprache)
The development of AI-driven computational methods has enabled rapid and non-invasive analysis of respiratory sounds using acoustic data, particularly cough recordings. Although the COVID-19 pandemic accelerated research on cough-based acoustic analysis, many early studies were limited by insufficient data quality, lack of standardized protocols, and limited reproducibility due to data scarcity. In this study, we propose an audio analysis framework for cough-based respiratory disease screening research using COVID-19 as a clinically validated case dataset. All analyses were conducted on a single clinically acquired multicentric dataset collected under standardized conditions in certified laboratories in Mexico and Spain, comprising cough recordings from 1105 individuals. Model training and testing were performed exclusively within this dataset. The framework incorporates signal preprocessing and a comparative evaluation of segmentation strategies, showing that segmented cough analysis significantly outperforms full-signal analysis. Class imbalance was addressed using the Synthetic Minority Over-sampling Technique (SMOTE) for CNN2D models and the supervised Resample filter implemented in WEKA for classical machine learning models, both applied exclusively to the training subset to generate balanced training sets and prevent data leakage. Feature extraction and classification were carried out using Random Forest, Support Vector Machine (SVM), XGBoost, and a 2D Convolutional Neural Network (CNN2D), with hyperparameter optimization via AutoML. The proposed framework achieved a best balanced screening performance of 85.58% sensitivity and 86.65% specificity (Random Forest with GeMAPSvB01), while the highest-specificity configuration reached 93.90% specificity with 18.14% sensitivity (CNN2D with SMOTE and AutoML). These results demonstrate the methodological feasibility of the proposed framework under the evaluated conditions.
Förderung
- National Council on Humanities, Sciences and Technologies
Zitieren
APA
Aquino-García, A. M., Pérez-Espinosa, H., Andreu-Perez, J., & González, A. Y. R. (2026). Audio-Based Screening of Respiratory Diseases Using Machine Learning: A Methodological Framework Evaluated on a Clinically Validated COVID-19 Cough Dataset. Machine Learning and Knowledge Extraction, 8(3), 80. https://doi.org/10.3390/make8030080
Vancouver
Aquino-García AM, Pérez-Espinosa H, Andreu-Perez J, González AYR. Audio-Based Screening of Respiratory Diseases Using Machine Learning: A Methodological Framework Evaluated on a Clinically Validated COVID-19 Cough Dataset. Machine Learning and Knowledge Extraction. 2026;8(3):80. doi:10.3390/make8030080
BibTeX
@article{AquinoGarca2026,
author = {Aquino-García, Arley Magnolia and Pérez-Espinosa, Humberto and Andreu-Perez, Javier and González, Ansel Y. Rodríguez},
title = {Audio-Based Screening of Respiratory Diseases Using Machine Learning: A Methodological Framework Evaluated on a Clinically Validated COVID-19 Cough Dataset},
journal = {Machine Learning and Knowledge Extraction},
year = {2026},
volume = {8},
number = {3},
pages = {80},
doi = {10.3390/make8030080}
}
EUVIMED