Feature Engineering and Deep Learning in Respiratory Sound Classification: A Systematic Literature Review
JOINTECS (Journal of Information Technology and Computer Science) · 2025
DOI: 10.31328/jointecs.v10i2.7411 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift JOINTECS (Journal of Information Technology and Computer Science), 2025. Verfasst von Mambaus Saadah und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
This systematic literature review examines research advancements in respiratory sound classification from 2020 to 2025, focusing on the integration of feature engineering and deep learning techniques. Thirty primary studies were analyzed to identify key trends in acoustic feature representation, model development, dataset utilization, and performance evaluation. MFCC, spectrograms, and emerging time–frequency representations such as scalograms and hybrid PSD–HHT features remain central to improving discriminative capability. Deep learning approaches demonstrate a clear methodological progression, transitioning from baseline CNN models to hybrid CNN–LSTM architectures and attention-based Transformer frameworks, which enhance temporal modeling, robustness, and interpretability. The ICBHI 2017 dataset continues to dominate empirical evaluations, though concerns regarding generalizability highlight the need for broader cross-dataset validation and real-world clinical testing. Overall, this review provides a consolidated synthesis of methodological strengths, limitations, and research gaps, offering a strategic roadmap for future investigations aimed at developing reliable, explainable, and clinically deployable respiratory sound classification systems.
Zitieren
APA
Saadah, M., & Zaeni, I. A. E. (2025). Feature Engineering and Deep Learning in Respiratory Sound Classification: A Systematic Literature Review. JOINTECS (Journal of Information Technology and Computer Science). https://doi.org/10.31328/jointecs.v10i2.7411
Vancouver
Saadah M, Zaeni IAE. Feature Engineering and Deep Learning in Respiratory Sound Classification: A Systematic Literature Review. JOINTECS (Journal of Information Technology and Computer Science). 2025. doi:10.31328/jointecs.v10i2.7411
BibTeX
@article{Saadah2025,
author = {Saadah, Mambaus and Zaeni, Ilham Ari Elbaith},
title = {Feature Engineering and Deep Learning in Respiratory Sound Classification: A Systematic Literature Review},
journal = {JOINTECS (Journal of Information Technology and Computer Science)},
year = {2025},
doi = {10.31328/jointecs.v10i2.7411}
}
EUVIMED