TY - GEN
T1 - DETECTION OF COVID-19 FROM JOINT TIME AND FREQUENCY ANALYSIS OF SPEECH, BREATHING AND COUGH AUDIO
AU - Harvill, John
AU - Wani, Yash
AU - Chatterjee, Moitreya
AU - Alam, Mustafa
AU - Beiser, David G.
AU - Chestek, David
AU - Hasegawa-Johnson, Mark
AU - Ahuja, Narendra
N1 - Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - The distinct cough sounds produced by a variety of respiratory diseases suggest the potential for the development of a new class of audio bio-markers for the detection of COVID-19. Accurate audio biomarker-based COVID-19 tests would be inexpensive, readily scalable, and non-invasive. Audio biomarker screening could also be utilized in resource-limited settings prior to traditional diagnostic testing. Here we explore the possibility of leveraging three audio modalities: cough, breathing, and speech to determine COVID-19 status. We train a separate neural classification system on each modality, as well as a fused classification system on all three modalities together. Ablation studies are performed to understand the relationship between individual and collective performance of the modalities. Additionally, we analyze the extent to which temporal and spectral features contribute to COVID-19 status information contained in the audio signals.
AB - The distinct cough sounds produced by a variety of respiratory diseases suggest the potential for the development of a new class of audio bio-markers for the detection of COVID-19. Accurate audio biomarker-based COVID-19 tests would be inexpensive, readily scalable, and non-invasive. Audio biomarker screening could also be utilized in resource-limited settings prior to traditional diagnostic testing. Here we explore the possibility of leveraging three audio modalities: cough, breathing, and speech to determine COVID-19 status. We train a separate neural classification system on each modality, as well as a fused classification system on all three modalities together. Ablation studies are performed to understand the relationship between individual and collective performance of the modalities. Additionally, we analyze the extent to which temporal and spectral features contribute to COVID-19 status information contained in the audio signals.
KW - COVID-19
KW - DiCOVA-II
KW - Telemedicine
UR - http://www.scopus.com/inward/record.url?scp=85131228506&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85131228506&partnerID=8YFLogxK
U2 - 10.1109/ICASSP43922.2022.9746015
DO - 10.1109/ICASSP43922.2022.9746015
M3 - Conference contribution
AN - SCOPUS:85131228506
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 3683
EP - 3687
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Y2 - 23 May 2022 through 27 May 2022
ER -