EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Interpretability analysis of deep models for COVID-19 detection

Daniel Peixoto Pinto da Silva, Edresson Casanova, Lucas Rafael Stefanel Gris, Marcelo Matheus Gauy, Arnaldo Candido Junior, Marcelo Finger, Flaviane Romani Fernandes Svartman, Beatriz Raposo de Medeiros, Marcus Vinícius Moreira Martins, Sandra Maria Aluísio, Larissa Cristina Berti, João Paulo Teixeira

Artificial Intelligence in Health · 2024

DOI: 10.36922/aih.2992 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Artificial Intelligence in Health, 2024. Verfasst von Daniel Peixoto Pinto da Silva und 11 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

During the coronavirus disease 2019 (COVID-19) pandemic, various research disciplines collaborated to address the impacts of severe acute respiratory syndrome coronavirus-2 infections. This paper presents an interpretability analysis of a convolutional neural network-based model designed for COVID-19 detection using audio data. We explore the input features that play a crucial role in the model’s decision-making process, including spectrograms, fundamental frequency (F0), F0 standard deviation, sex, and age. Subsequently, we examine the model’s decision patterns by generating heat maps to visualize its focus during the decision-making process. Emphasizing an explainable artificial intelligence approach, our findings demonstrate that the examined models can make unbiased decisions even in the presence of noise in training set audios, provided appropriate preprocessing steps are undertaken. Our top-performing model achieves a detection accuracy of 94.44%. Our analysis indicates that the analyzed models prioritize high-energy areas in spectrograms during the decision process, particularly focusing on high-energy regions associated with prosodic domains, while also effectively utilizing F0 for COVID-19 detection.

Zitieren

APA

Silva, D. P. P. D., Casanova, E., Gris, L. R. S., Gauy, M. M., Junior, A. C., Finger, M., Svartman, F. R. F., Medeiros, B. R. D., Martins, M. V. M., Aluísio, S. M., Berti, L. C., & Teixeira, J. P. (2024). Interpretability analysis of deep models for COVID-19 detection. Artificial Intelligence in Health. https://doi.org/10.36922/aih.2992

Vancouver

Silva DPPD, Casanova E, Gris LRS, Gauy MM, Junior AC, Finger M, et al. Interpretability analysis of deep models for COVID-19 detection. Artificial Intelligence in Health. 2024. doi:10.36922/aih.2992

BibTeX
@article{Silva2024,
  author = {Silva, Daniel Peixoto Pinto da and Casanova, Edresson and Gris, Lucas Rafael Stefanel and Gauy, Marcelo Matheus and Junior, Arnaldo Candido and Finger, Marcelo and Svartman, Flaviane Romani Fernandes and Medeiros, Beatriz Raposo de and Martins, Marcus Vinícius Moreira and Aluísio, Sandra Maria and Berti, Larissa Cristina and Teixeira, João Paulo},
  title = {Interpretability analysis of deep models for COVID-19 detection},
  journal = {Artificial Intelligence in Health},
  year = {2024},
  doi = {10.36922/aih.2992}
}
Kostenlos starten →