EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Comparative Evaluation of MFCC and Mel-spectrogram Features for CNN-Based Respiratory Abnormality Detection

Simboni Simboni Tege, Kafunda Katalay Pierre, Oshasha Oshasha Fiston, Sylvestre Frey, Albert Ntumba Nkongolo, Biaba Kuya Jirince

Journal of Applied Informatics and Computing · 16.04.2026 · 10 (2) · 1142-1150

DOI: 10.30871/jaic.v10i2.12355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Applied Informatics and Computing, online erschienen am 16.04.2026. Verfasst von Simboni Simboni Tege und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Automated respiratory sound analysis addresses critical limitations in traditional clinical auscultation, particularly high inter-observer variability and limited specialist access in resource-constrained settings. This study rigorously compares Mel-Frequency Cepstral Coefficients (MFCC) and Mel-spectrogram representations for classifying respiratory abnormalities using convolutional neural networks. Using the ICBHI 2017 dataset (920 recordings, 6,898 cycles from 126 patients), we implemented identical CNN architectures differing only in input features. Class imbalance was addressed through Synthetic Minority Over-sampling Technique applied exclusively to training data. The MFCC model achieved 83% accuracy with superior sensitivity for normal sounds (97% recall), while Mel-spectrograms reached 82% accuracy with higher precision (95%). MFCC demonstrated better crackle detection (76% vs 73% recall) and wheeze precision (75% vs 71%), attributed to enhanced transient spectral capture through discrete cosine transformation. Both models showed strong discrimination (AUC > 0.90). MFCC offers computational efficiency advantages for screening applications, while Mel-spectrograms provide interpretability for diagnostic contexts. This controlled comparison provides evidence-based guidance for computer-aided respiratory diagnostic system design, particularly relevant for resource-limited healthcare environments.

Zitieren

APA

Tege, S. S., Pierre, K. K., Fiston, O. O., Frey, S., Nkongolo, A. N., & Jirince, B. K. (2026). Comparative Evaluation of MFCC and Mel-spectrogram Features for CNN-Based Respiratory Abnormality Detection. Journal of Applied Informatics and Computing, 10(2), 1142-1150. https://doi.org/10.30871/jaic.v10i2.12355

Vancouver

Tege SS, Pierre KK, Fiston OO, Frey S, Nkongolo AN, Jirince BK. Comparative Evaluation of MFCC and Mel-spectrogram Features for CNN-Based Respiratory Abnormality Detection. Journal of Applied Informatics and Computing. 2026;10(2):1142-1150. doi:10.30871/jaic.v10i2.12355

BibTeX
@article{Tege2026,
  author = {Tege, Simboni Simboni and Pierre, Kafunda Katalay and Fiston, Oshasha Oshasha and Frey, Sylvestre and Nkongolo, Albert Ntumba and Jirince, Biaba Kuya},
  title = {Comparative Evaluation of MFCC and Mel-spectrogram Features for CNN-Based Respiratory Abnormality Detection},
  journal = {Journal of Applied Informatics and Computing},
  year = {2026},
  volume = {10},
  number = {2},
  pages = {1142--1150},
  doi = {10.30871/jaic.v10i2.12355}
}
Kostenlos starten →