An Ensemble Approach for the Diagnosis of COVID-19 from Speech and Cough Sounds

Fathan, Abderrahim; Alam, Jahangir; Kang, Woo Hyun

doi:10.1007/978-3-030-87802-3_18

Cited by 2 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The previous literature has shown that classification of COVID-19 using acoustic signatures is indeed possible: Laguarta et al [ 9 ] achieved a 93.8% accuracy on forced-cough recordings with parallel ResNet50 deep learning architectures; Imran et al [ 10 ] used three parallel classifier systems with a mediator to achieve a final accuracy of 92.64% (though the app predicted an inconclusive test result 38.7% of the time, which was not accounted for in the accuracy); Pahar et al [ 11 ] applied transfer learning on a pre-trained ResNet50 architecture to achieve accuracies above 92% for cough, speech, and breathing sounds; and Pinkas et al [ 12 ] used a three stage deep learning architecture to correctly identify 71% of positive patients. The release of public datasets, such as Coswara/DiCOVA Challenge [ 13 , 14 ], University of Cambridge/NeurIPS 2021 [ 15 ], and COUGHVID [ 16 ] has dramatically accelerated the development and release of new classification approaches with reported area-under-the-curve of the receiver operating curve (AUC-ROC) ranging from 0.60 to 0.95 [ 17 , 18 , 19 , 20 ]. Previously, the authors have also presented early work on the Coswara Dataset [ 21 ] that was the top performer in the breathing and cough tracks of the Second DiCOVA Challenge, achieving an AUC-ROC of 0.87 and 0.82, respectively [ 22 ].…”

Section: Introductionmentioning

confidence: 99%

Considerations and Challenges for Real-World Deployment of an Acoustic-Based COVID-19 Screening System

Grant

McLane

Rennoll

et al. 2022

Sensors

View full text Add to dashboard Cite

Coronavirus disease 2019 (COVID-19) has led to countless deaths and widespread global disruptions. Acoustic-based artificial intelligence (AI) tools could provide a simple, scalable, and prompt method to screen for COVID-19 using easily acquirable physiological sounds. These systems have been demonstrated previously and have shown promise but lack robust analysis of their deployment in real-world settings when faced with diverse recording equipment, noise environments, and test subjects. The primary aim of this work is to begin to understand the impacts of these real-world deployment challenges on the system performance. Using Mel-Frequency Cepstral Coefficients (MFCC) and RelAtive SpecTrAl-Perceptual Linear Prediction (RASTA-PLP) features extracted from cough, speech, and breathing sounds in a crowdsourced dataset, we present a baseline classification system that obtains an average receiver operating characteristic area under the curve (AUC-ROC) of 0.77 when discriminating between COVID-19 and non-COVID subjects. The classifier performance is then evaluated on four additional datasets, resulting in performance variations between 0.64 and 0.87 AUC-ROC, depending on the sound type. By analyzing subsets of the available recordings, it is noted that the system performance degrades with certain recording devices, noise contamination, and with symptom status. Furthermore, performance degrades when a uniform classification threshold from the training data is subsequently used across all datasets. However, the system performance is robust to confounding factors, such as gender, age group, and the presence of other respiratory conditions. Finally, when analyzing multiple speech recordings from the same subjects, the system achieves promising performance with an AUC-ROC of 0.78, though the classification does appear to be impacted by natural speech variations. Overall, the proposed system, and by extension other acoustic-based diagnostic aids in the literature, could provide comparable accuracy to rapid antigen testing but significant deployment challenges need to be understood and addressed prior to clinical use.

show abstract

Section: Introductionmentioning

confidence: 99%

Considerations and Challenges for Real-World Deployment of an Acoustic-Based COVID-19 Screening System

Grant

McLane

Rennoll

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

An application development for smart monitoring of COVID patients using six stage microbiological health systems

Prasad¹,

Nair²,

Patil³

et al. 2023

IFS

View full text Add to dashboard Cite

For many, Covid-19 is a short-term, mildly debilitating disease. But some people are still struggling with monthly symptoms with persistent inflammation, chronic pain and shortness of breath. The situation of “long-term cowardice” has become so debilitating that it is now common for some to say that they are tired even if they walk a short distance. So far, the focus has been on saving lives from the plague. But now there are growing concerns about people facing the long-term consequences of the COVID epidemic. The fundamental question, with the uncertainty of whether those with chronic goiter, or all those affected, will fully recover is raised. In this paper a smart monitoring model was proposed to keep monitoring the COVID patient’s health conditions. The smart method keep on watching the different changes reflected in the body conditions and ensure the changes in the database. In case any emergency is raised, then these smart monitoring tools inform the information to the doctors. This can very much helpful for the patients to communicate with the doctors.

show abstract

An Ensemble Approach for the Diagnosis of COVID-19 from Speech and Cough Sounds

Cited by 2 publications

References 19 publications

Considerations and Challenges for Real-World Deployment of an Acoustic-Based COVID-19 Screening System

Considerations and Challenges for Real-World Deployment of an Acoustic-Based COVID-19 Screening System

An application development for smart monitoring of COVID patients using six stage microbiological health systems

Contact Info

Product

Resources

About