A Generic Deep Learning Based Cough Analysis System From Clinically Validated Samples for Point-of-Need Covid-19 Test and Severity Levels

Abstract

In an attempt to reduce the infection rate of the COrona VIrus Disease-19 (Covid-19) countries around the world have echoed the exigency for an economical, accessible, point-of-need diagnostic test to identify Covid-19 carriers so that they (individuals who test positive) can be advised to self isolate rather than the entire community. Availability of a quick turnaround time diagnostic test would essentially mean that life, in general, can return to normality-at-large. In this regards, studies concurrent in time with ours have investigated different respiratory sounds, including cough, to recognise potential Covid-19 carriers. However, these studies lack clinical control and rely on Internet users confirming their test results in a web questionnaire (crowdsourcing) thus rendering their analysis inadequate. We seek to evaluate the detection performance of a primary screening tool of Covid-19 solely based on the cough sound from 8,380 clinically validated samples with laboratory molecular-test (2,339 Covid-19 positive and 6,041 Covid-19 negative) under quantitative RT-PCR (qRT-PCR) from certified laboratories. All collected samples were clinically labelled, i.e. Covid-19 positive or negative, according to the results in addition to the disease severity based on the qRT-PCR threshold cycle (Ct) and lymphocytes count from the patients. Our proposed generic method is an algorithm based on Empirical Mode Decomposition (EMD) for cough sound detection with subsequent classification based on a tensor of audio sonographs and deep artificial neural network classifier with convolutional layers called 'DeepCough'. Two different versions of DeepCough based on the number of tensor dimensions, i.e. DeepCough2D and DeepCough3D, have been investigated. These methods have been deployed in a multi-platform prototype web-app 'CoughDetect'.Covid-19 recognition results rates achieved a promising AUC (Area Under Curve) of 98.80% ± 0.83%, sensitivity of 96.43% ± 1.85%, and specificity of 96.20% ± 1.74% and average AUC of 81.08% ± 5.05% for the recognition of three severity levels. Our proposed web tool as a pointof-need primary diagnostic test for Covid-19 facilitates the rapid detection of the infection. We believe it has the potential to significantly hamper the Covid-19 pandemic across the world.

COVID-19 detection from audio: seven grains of salt

“…However, the validity of the labels used in the COVID-19 audio datasets collected so far is questionable because most datasets allow participants to self-report their COVID-19 status and fail to record the type of test participants had; 2 , 8 , 9 polymerase chain reaction (PCR) or, the less accurate, lateral flow test. Although some studies do demand a validated PCR test, 6 , 7 , 8 , 10 the datasets are small and none have been made publicly available at the time of writing. To highlight the severity of this problem, we note that some datasets have accepted self-assessment as a means of labelling the dataset, 3 and others have failed even to detail how COVID-19 status was determined.…”

“…When participant populations are not controlled, inflated classification scores are reported because the model can easily recognise reappearing participants and classify their COVID-19 status based on cases in the training phase. Nevertheless, several datasets do not record the identity of participants, 10 , 9 resulting in an inability to avert this issue.…”

COVID-19 cough classification using machine learning and global smartphone recordings

“…COVID-19 coughs were classified with a higher AUC of 0.97 (sensitivity = 98.5% and specificity = 94.2%) by a Resnet50 architecture, trained on coughs from 4256 subjects and evaluated on 1064 subjects that included both COVID-19 positive and COVID-19 negative subjects by implementing four biomarkers [ 28 ]. A high AUC exceeding 0.98 was also achieved in [ 29 ] when discriminating COVID-19 positive coughs from COVID-19 negative coughs on a clinically validated dataset consisting of 2339 COVID-19 positive and 6041 COVID-19 negative subjects using DNN based classifiers.…”

Applications of Small and Medium Enterprise Management System Using Edge Algorithm

“…e indirect degree of interest between information also reflects the similarity between information to a certain extent. It can be simply considered that the user's indirect score of information represents the user's indirect interest in information [18]. e indirect score of the user υ to information i is defined as follows: if a user u has a rating on information i, the user υ does not store information i, but the user υ is similar to the user u; then it can be considered that the user υ has an indirect score on information i through a user u. e direct interest vector InD 1 , InD 2 , .…”