Acoustic surveillance of cough for detecting respiratory disease using artificial intelligence

Abstract

Research questionCan smartphones be used to detect individual and population-level changes in cough frequency that correlate with the incidence of COVID-19 and other respiratory infections?MethodsThis was a prospective cohort study carried out in Pamplona (Spain) between 2020 and 2021 using artificial intelligence cough detection software. Changes in cough frequency around the time of medical consultation were evaluated using a randomisation routine, significance was tested by comparing the distribution of cough frequencies to that obtained from a model of no difference. The correlation between changes of cough frequency and COVID-19 incidence was studied using an autoregressive moving average (ARIMA) analysis, and its strength determined by calculating its auto-correlation function (ACF). Predictors for the regular use of the system were studied using a linear regression. Overall user experience was evaluated with a satisfaction questionnaire and through focused group discussions.ResultsWe followed up 616 participants and collected over 62 000 coughs. Coughs per hour surged around the time cohort subjects sought medical care (difference=+0.77 coughs h−1, p=0.00001) There was a weak temporal correlation between aggregated coughs and the incidence of COVID-19 in the local population (ACF=0.43). Technical issues affected uptake and regular use of the system.InterpretationArtificial intelligence systems can detect changes in cough frequency that temporarily correlate with the onset of clinical disease at the individual level. A clearer correlation with population-level COVID-19 incidence, or other respiratory conditions, could be achieved with better penetration and compliance with cough monitoring.

Digital cough monitoring – A potential predictive acoustic biomarker of clinical outcomes in hospitalized COVID-19 patients

“…This CNN model was previously shown to have above 96·0% analytical sensitivity and specificity and was re-validated using gold-standard human sound labelling within this study’s specific setting. [7] , [8] , [9] Phones were positioned at the head of hospital beds with microphones oriented towards the patients within three feet of the patients’ mouths. Phones were left charging with the Hyfe Research application activated until the completion of their enrolment period.…”

The Statistics of Counting Coughs: Easy as 1, 2, 3?

“…The first step towards a better understanding of cough is diligent data collection: track all coughs as they occur and maintain records that can be carefully analyzed [6,7]. What we are learning is that coughs are like traffic on city streets.…”

Performance evaluation of the smartphone-based AI cough monitoring app - Hyfe Cough Tracker against solicited respiratory sounds

“…Recent advances in artificial intelligence (AI) allow the monitoring of cough in a non-obtrusive way using smartphones or other wearable digital devices. 6,[11][12][13][14] Unobtrusive and privacy preserving passive cough monitors could revolutionize clinical practice and research in the field of respiratory diseases.…”

“…Longitudinal cough monitoring also opens the door to population-wide capture of cough-signals as a surrogate marker of respiratory diseases epidemiology. 14 Evaluating cough and its patterns with limited recording periods (e.g., 24 h) can be misleading, in particular, if only small changes in cough frequency are captured over the limited 24 h recording and in cases that have high variance of cough counts. 12 However, the nature and volume of data generated with protracted monitoring raises new challenges in technology validation.…”