Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection

Ciro Rodriguez Rodriguez; Daniel Angeles; Renzo Chafloque; Freddy Kaseng; Bishwajeet Pandey

doi:10.1109/CICN49253.2020.9242583

Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection

Ciro Rodriguez Rodriguez, Daniel Angeles, Renzo Chafloque, Freddy Kaseng, Bishwajeet Pandey

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

18 Scopus citations

Abstract

The objective of the paper is to provide a model capable of serving as a basis for retraining a convolutional neural network that can be used to detect COVID-19 cases through spectrograms of coughing, sneezing and other respiratory sounds from infected people. To address this challenge, the methodology was focused on Deep Learning technics worked with a dataset of sounds of sick and non-sick people, and using ImageNet's Xception architecture to train the model to be presented through Fine-Tuning. The results obtained were a precision of 0.75 to 0.80, this being drastically affected by the quality of the dataset at our availability, however, when getting relatively high results for the conditions of the data used, we can conclude that the model can present much better results if it is working with a dataset specifically of respiratory sounds of COVID-19 cases with high quality.

Original language	English
Title of host publication	Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020
Editors	Geetam Tomar
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	429-434
Number of pages	6
ISBN (Electronic)	9781728193939
DOIs	https://doi.org/10.1109/CICN49253.2020.9242583
State	Published - 25 Sep 2020
Event	12th International Conference on Computational Intelligence and Communication Networks, CICN 2020 - Bhimtal, India Duration: 25 Sep 2020 → 26 Sep 2020

Publication series

Name	Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020

Conference

Conference	12th International Conference on Computational Intelligence and Communication Networks, CICN 2020
Country/Territory	India
City	Bhimtal
Period	25/09/20 → 26/09/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Keywords

COVID-19
ImageNet's Xception architecture
audio spectrograms
convolutional neural networks
deep learning

Access to Document

10.1109/CICN49253.2020.9242583

Cite this

Rodriguez, C. R., Angeles, D., Chafloque, R., Kaseng, F., & Pandey, B. (2020). Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection. In G. Tomar (Ed.), Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020 (pp. 429-434). Article 9242583 (Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CICN49253.2020.9242583

Rodriguez, Ciro Rodriguez ; Angeles, Daniel ; Chafloque, Renzo et al. / Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection. Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020. editor / Geetam Tomar. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 429-434 (Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020).

@inproceedings{79af17f264ab4b82a9f43fea2b3306cf,

title = "Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection",

abstract = "The objective of the paper is to provide a model capable of serving as a basis for retraining a convolutional neural network that can be used to detect COVID-19 cases through spectrograms of coughing, sneezing and other respiratory sounds from infected people. To address this challenge, the methodology was focused on Deep Learning technics worked with a dataset of sounds of sick and non-sick people, and using ImageNet's Xception architecture to train the model to be presented through Fine-Tuning. The results obtained were a precision of 0.75 to 0.80, this being drastically affected by the quality of the dataset at our availability, however, when getting relatively high results for the conditions of the data used, we can conclude that the model can present much better results if it is working with a dataset specifically of respiratory sounds of COVID-19 cases with high quality.",

keywords = "COVID-19, ImageNet's Xception architecture, audio spectrograms, convolutional neural networks, deep learning",

author = "Rodriguez, {Ciro Rodriguez} and Daniel Angeles and Renzo Chafloque and Freddy Kaseng and Bishwajeet Pandey",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020 ; Conference date: 25-09-2020 Through 26-09-2020",

year = "2020",

month = sep,

day = "25",

doi = "10.1109/CICN49253.2020.9242583",

language = "Ingl{\'e}s",

series = "Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "429--434",

editor = "Geetam Tomar",

booktitle = "Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020",

address = "Estados Unidos",

}

Rodriguez, CR, Angeles, D, Chafloque, R, Kaseng, F & Pandey, B 2020, Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection. in G Tomar (ed.), Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020., 9242583, Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020, Institute of Electrical and Electronics Engineers Inc., pp. 429-434, 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020, Bhimtal, India, 25/09/20. https://doi.org/10.1109/CICN49253.2020.9242583

Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection. / Rodriguez, Ciro Rodriguez; Angeles, Daniel; Chafloque, Renzo et al.
Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020. ed. / Geetam Tomar. Institute of Electrical and Electronics Engineers Inc., 2020. p. 429-434 9242583 (Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection

AU - Rodriguez, Ciro Rodriguez

AU - Angeles, Daniel

AU - Chafloque, Renzo

AU - Kaseng, Freddy

AU - Pandey, Bishwajeet

PY - 2020/9/25

Y1 - 2020/9/25

N2 - The objective of the paper is to provide a model capable of serving as a basis for retraining a convolutional neural network that can be used to detect COVID-19 cases through spectrograms of coughing, sneezing and other respiratory sounds from infected people. To address this challenge, the methodology was focused on Deep Learning technics worked with a dataset of sounds of sick and non-sick people, and using ImageNet's Xception architecture to train the model to be presented through Fine-Tuning. The results obtained were a precision of 0.75 to 0.80, this being drastically affected by the quality of the dataset at our availability, however, when getting relatively high results for the conditions of the data used, we can conclude that the model can present much better results if it is working with a dataset specifically of respiratory sounds of COVID-19 cases with high quality.

AB - The objective of the paper is to provide a model capable of serving as a basis for retraining a convolutional neural network that can be used to detect COVID-19 cases through spectrograms of coughing, sneezing and other respiratory sounds from infected people. To address this challenge, the methodology was focused on Deep Learning technics worked with a dataset of sounds of sick and non-sick people, and using ImageNet's Xception architecture to train the model to be presented through Fine-Tuning. The results obtained were a precision of 0.75 to 0.80, this being drastically affected by the quality of the dataset at our availability, however, when getting relatively high results for the conditions of the data used, we can conclude that the model can present much better results if it is working with a dataset specifically of respiratory sounds of COVID-19 cases with high quality.

KW - COVID-19

KW - ImageNet's Xception architecture

KW - audio spectrograms

KW - convolutional neural networks

KW - deep learning

UR - http://www.scopus.com/inward/record.url?scp=85096827603&partnerID=8YFLogxK

U2 - 10.1109/CICN49253.2020.9242583

DO - 10.1109/CICN49253.2020.9242583

M3 - Contribución a la conferencia

AN - SCOPUS:85096827603

T3 - Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020

SP - 429

EP - 434

BT - Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020

A2 - Tomar, Geetam

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020

Y2 - 25 September 2020 through 26 September 2020

ER -

Rodriguez CR, Angeles D, Chafloque R, Kaseng F, Pandey B. Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection. In Tomar G, editor, Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 429-434. 9242583. (Proceedings - 2020 12th International Conference on Computational Intelligence and Communication Networks, CICN 2020). doi: 10.1109/CICN49253.2020.9242583

Deep Learning Audio Spectrograms Processing to the Early COVID-19 Detection

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this