Discovery and Analytical Validation of a Vocal Biomarker to Monitor Anosmia and Ageusia in Patients With COVID-19: Cross-sectional Study

Abstract

Background The COVID-19 disease has multiple symptoms, with anosmia and ageusia being the most prevalent, varying from 75% to 95% and from 50% to 80% of infected patients, respectively. An automatic assessment tool for these symptoms will help monitor the disease in a fast and noninvasive manner. Objective We hypothesized that people with COVID-19 experiencing anosmia and ageusia had different voice features than those without such symptoms. Our objective was to develop an artificial intelligence pipeline to identify and internally validate a vocal biomarker of these symptoms for remotely monitoring them. Methods This study used population-based data. Participants were assessed daily through a web-based questionnaire and asked to register 2 different types of voice recordings. They were adults (aged >18 years) who were confirmed by a polymerase chain reaction test to be positive for COVID-19 in Luxembourg and met the inclusion criteria. Statistical methods such as recursive feature elimination for dimensionality reduction, multiple statistical learning methods, and hypothesis tests were used throughout this study. The TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) Prediction Model Development checklist was used to structure the research. Results This study included 259 participants. Younger (aged <35 years) and female participants showed higher rates of ageusia and anosmia. Participants were aged 41 (SD 13) years on average, and the data set was balanced for sex (female: 134/259, 51.7%; male: 125/259, 48.3%). The analyzed symptom was present in 94 (36.3%) out of 259 participants and in 450 (27.5%) out of 1636 audio recordings. In all, 2 machine learning models were built, one for Android and one for iOS devices, and both had high accuracy—88% for Android and 85% for iOS. The final biomarker was then calculated using these models and internally validated. Conclusions This study demonstrates that people with COVID-19 who have anosmia and ageusia have different voice features from those without these symptoms. Upon further validation, these vocal biomarkers could be nested in digital devices to improve symptom assessment in clinical practice and enhance the telemonitoring of COVID-19–related symptoms. Trial Registration Clinicaltrials.gov NCT04380987; https://clinicaltrials.gov/ct2/show/NCT04380987

Co-Design of a Voice-Based Digital Health Solution to Monitor Persisting Symptoms Related to COVID-19 (UpcomingVoice Study): Protocol for a Mixed Methods Study

“…To date, almost 6000 voice recordings from more than >500 patients with COVID-19 have already been collected in the Predi-COVID study [2]. These voice recordings have been analyzed, and promising vocal biomarker candidates for fatigue, loss of taste and smell, and symptomatic status in people with a COVID-19 infection have already been identified [20][21][22].…”

Current Practices in Voice Data Collection and Limitations to Voice AI Research: A National Survey

“…Additionally, the COVID-19 pandemic generated increased interest in developing technology to perform remote, minimally invasive respiratory disease diagnostics, and voicederived biomarkers have come to the forefront of this effort. [3][4][5][6][7] Many groups have also found grounds for the analysis of voice for the detection and progression of neurological disorders, such as Parkinson's, Alzheimer's, Huntington's, and dementia. [8][9][10][11][12][13][14][15][16] Additional areas of investigation include pediatric and adult voice disorders, and mental health disorders.…”

“…The vocal biomarkers market was worth USD 1901.1 million in 2021 and is expected to reach USD 5131.8 million in 2028 2 . Additionally, the COVID‐19 pandemic generated increased interest in developing technology to perform remote, minimally invasive respiratory disease diagnostics, and voice‐derived biomarkers have come to the forefront of this effort 3–7 . Many groups have also found grounds for the analysis of voice for the detection and progression of neurological disorders, such as Parkinson's, Alzheimer's, Huntington's, and dementia 8–16 .…”

Voice as a Biomarker in Health‐Tech: Mapping the Evolving Landscape of Voice Biomarkers in the Start‐Up World

“…Speech is the articulation of words and sentences as a primary mode of expressing thoughts, ideas, emotions, or information between individuals. The biological information derived from voice and speech can be utilized for the development of “vocal biomarkers”—objective acoustic parameters, linguistics, and paralinguistics markers to be applied to the screening, diagnosis, and monitoring of human health 3‐7 …”

“…The biological information derived from voice and speech can be utilized for the development of "vocal biomarkers"-objective acoustic parameters, linguistics, and paralinguistics markers to be applied to the screening, diagnosis, and monitoring of human health. [3][4][5][6][7] The earliest implementation of voice analysis in health occurred by way of manual cough counting in 1988. 8 Since then, the advancement of digital technology, particularly artificial intelligence (AI) and machine learning, has facilitated automation and further exploration of the field, expanding the use case into screening and monitoring of a variety of disorders.…”