End-to-end convolutional neural network enables COVID-19 detection from breath and cough audio: a pilot study

Abstract

BackgroundSince the emergence of COVID-19 in December 2019, multidisciplinary research teams have wrestled with how best to control the pandemic in light of its considerable physical, psychological and economic damage. Mass testing has been advocated as a potential remedy; however, mass testing using physical tests is a costly and hard-to-scale solution.MethodsThis study demonstrates the feasibility of an alternative form of COVID-19 detection, harnessing digital technology through the use of audio biomarkers and deep learning. Specifically, we show that a deep neural network based model can be trained to detect symptomatic and asymptomatic COVID-19 cases using breath and cough audio recordings.ResultsOur model, a custom convolutional neural network, demonstrates strong empirical performance on a data set consisting of 355 crowdsourced participants, achieving an area under the curve of the receiver operating characteristics of 0.846 on the task of COVID-19 classification.ConclusionThis study offers a proof of concept for diagnosing COVID-19 using cough and breath audio signals and motivates a comprehensive follow-up research study on a wider data sample, given the evident advantages of a low-cost, highly scalable digital COVID-19 diagnostic tool.

CovNet: A Transfer Learning Framework for Automatic COVID-19 Detection From Crowd-Sourced Cough Sounds

“…In Hershey et al's ( 48 ) study,ResNet has outperformed other CNNs for audio classification on the Audio Set ( 49 ). A ResNet based model is constructed for COVID-19 detection from breath and cough audio signals ( 50 ).…”

Clinical and Laboratory Approach to Diagnose COVID-19 Using Machine Learning

“…High cost and availability of CT-Scanners is also an issue. Research has also taken place in exploring the use of cough sounds for COVID-19 diagnosis using NLP (Natural Language Processing) [26][27][28].…”

AUCO ResNet: an end-to-end network for Covid-19 pre-screening from cough and breath

“…Taking into consideration the high percentage of asymptomatic yet contagious COVID-19 patients, researchers have argued that forced-cough (i.e. voluntary cough [11] ) keeps the same biomarker potential of the spontaneous one, as data have wildly proved [ [10] , [11] , [12] , [14] , [15] , [16] , [17] , [18] , [19] ] . In fact, half of the asymptomatic cases present CT abnormalities [3] .…”

“…Thus a single neural network framework is capable of performing feature extraction, feature selection, dimensionality reduction and prediction by optimizing each single module. Other works such as [ 9 , 16 , 19 ] used end-to-end training and deep neural networks, however these works lack the optimization of the spectrogram's filter generation as an internal process. Moreover, no attention mechanisms are used to extract relevant patterns.…”