Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Nov:138:104944.
doi: 10.1016/j.compbiomed.2021.104944. Epub 2021 Oct 13.

Detection of COVID-19 from voice, cough and breathing patterns: Dataset and preliminary results

Vladimir Despotovic  1 , Muhannad Ismael  2 , Maël Cornil  2 , Roderick Mc Call  2 , Guy Fagherazzi  3
Affiliations

Detection of COVID-19 from voice, cough and breathing patterns: Dataset and preliminary results

Vladimir Despotovic et al. Comput Biol Med. 2021 Nov.

Abstract

COVID-19 heavily affects breathing and voice and causes symptoms that make patients' voices distinctive, creating recognizable audio signatures. Initial studies have already suggested the potential of using voice as a screening solution. In this article we present a dataset of voice, cough and breathing audio recordings collected from individuals infected by SARS-CoV-2 virus, as well as non-infected subjects via large scale crowdsourced campaign. We describe preliminary results for detection of COVID-19 from cough patterns using standard acoustic features sets, wavelet scattering features and deep audio embeddings extracted from low-level feature representations (VGGish and OpenL3). Our models achieve accuracy of 88.52%, sensitivity of 88.75% and specificity of 90.87%, confirming the applicability of audio signatures to identify COVID-19 symptoms. We furthermore provide an in-depth analysis of the most informative acoustic features and try to elucidate the mechanisms that alter the acoustic characteristics of coughs of people with COVID-19.

Keywords: Artificial intelligence; COVID-19; Cough; Digital biomarker; Voice.

PubMed Disclaimer

Figures

Fig. 1
Fig. 1
Distribution of symptoms across participants in CDCVA dataset.
Fig. 2
Fig. 2
Distribution of comorbidities across participants in CDCVA dataset DM: Diabetes Mellitus; COPD: Chronic obstructive pulmonary disease; PHTN: Pulmonary hypertension; GERD: Gastroesophageal reflux disease; LPR: Laryngopharyngeal reflux; PD: Parkinson's disease; MS: Multiple sclerosis; Neck scars: Scars from neck surgery or from trauma to the front of the neck.
Fig. 3
Fig. 3
The waveforms and the spectrograms of the cough signal a) before and b) after noise reduction.
Fig. 4
Fig. 4
Two-dimensional graph for selected features generated by t-distributed Stochastic Neighbor Embedding approach. Features are extracted using ComParE feature set and selected using mutual information criterion.
Fig. 5
Fig. 5
The list of 10 most informative features in ComParE acoustic feature set based on the mutual information criterion.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Sanders J., Monogue M., Jodlowski T., Cutrell J. Pharmacologic treatments for coronavirus disease 2019 (COVID-19): a review. J. Am. Med. Assoc. 2020;323(18):1824–1836. - PubMed
    1. Kujawski S., Wong K., Collins J., et al. Clinical and virologic characteristics of the first 12 patients with coronavirus disease 2019 (COVID-19) in the United States. Nat. Med. 2020;26:861–868. - PubMed
    1. Chang D., Lin M., Wei L., Xie L., Zhu G., Cruz C.S.D., Sharma L. Epidemiologic and clinical characteristics of novel coronavirus infections involving 13 patients outside Wuhan, China. J. Am. Med. Assoc. 2020;323(11):1092–1093. - PMC - PubMed
    1. Sun P., Lu X., Xu C., Sun W., Pan B. Understanding of COVID-19 based on current evidence. J. Med. Virol. 2020;92(6):548–551. - PMC - PubMed
    1. Asiaee M., Vahedian-azimi A., Atashi S.S., Keramatfar A., Nourbakhsh M. Voice quality evaluation in patients with COVID-19: an acoustic analysis. J. Voice. 2021 In press. - PMC - PubMed

Publication types