News • Vocal biomarker

Covid-19 'voice check' screening tool validated in large clinical study

AI-based vocal biomarker company Vocalis Health announced results of a clinical study conducted in collaboration with the Municipal Corporation of Greater Mumbai (MCGM) at their NESCO Covid-19 Center to validate Vocalis Health’s Covid-19 screening tool, VocalisCheck.

The study included over 2,000 participants who spoke numerous languages including English, Hindi, Marathi and Gujarati.  Results from an unblinded validation set of 288 participants demonstrated an AUC of 0.88, which translated to accuracy of 81.2 percent, sensitivity of 80.3 percent and specificity of 81.4 percent.

Image source: Pixabay/Mary Theresa McLean / CDC (Mashup: HiE)

In addition to the results of this study, Vocalis Health today announced that its Covid-19 screening tool, VocalisCheck, has obtained a CE mark. VocalisCheck is the first device with CE mark approval for its medical intended use of screening for Covid-19. “PCR testing is being used to screen for Covid-19, which is an extremely expensive, resource-intensive and time-consuming approach,” said Dr. Shady Hassan, Co-Founder, Chief Operating and Medical Officer of Vocalis Health. “Instead of misusing PCR testing for screening, utilizing a highly scalable screening tool like VocalisCheck can fill a significant gap in the current approach to Covid screening, with the ability to effectively funnel those with high risk of infection to the appropriate diagnostic test.”

VocalisCheck has now proven its reliability to effectively identify Covid-19 risk in multiple studies, including this study in India, delivering reliable risk assessment for Covid-19 infection even in people who are not symptomatic. In the study in India, a symptom checker alone correctly identified approximately 66 percent of the people with Covid-19 infection, while Vocalis Health’s Covid-19 vocal biomarker achieved >80 percent accuracy in assessing a person’s risk of Covid-19 infection. “We are encouraged by the study’s findings, which further validate VocalisCheck’s ability to screen for Covid-19,” explained Tal Wenderow, CEO of Vocalis Health. “These new results, combined with our recent CE-mark approval, demonstrate our commercial readiness to deploy the VocalisCheck screening tool to help businesses, governments, universities and others safely return to work, school, healthcare and leisure while lightening the burden on health systems.”

How does it work?

Photo
A single voice sample is recorded using a mobile app or web-based tool, and a cloud-based AI analytics engine provides feedback within a minute or two.

Screenshot: Vocalis Health

VocalisCheck is a software-only product that can be accessed via smartphones or other devices. Users simply count from 50 to 70, their voice recording is transformed to a picture (spectrogram) containing 512 features. That picture is then compared using AI machine learning/deep learning techniques to a composite picture based upon the voice recordings of Covid-19 positive participants in clinical trials to determine whether there is a high or low correlation between them.  This technology can be integrated into existing mobile or web applications (via an SDK or Software Developers Kit) or can be accessed through a dedicated web-based portal.

VocalisCheck is not intended to serve as a diagnostic tool, the company states.  It is designed for use as a screening tool that can deliver a reliable risk score to guide who may need additional diagnostic testing, thereby enabling health systems to allocate diagnostic resources more efficiently.


Source: Vocalis Health

18.02.2021

Read all latest stories

Related articles

Photo

News • Machine learning approach for screening

AI model predicts results of Covid-19 tests

A new study demonstrates that AI models, using symptom and demographic features, can help predict Covid-19 infections, providing a way for rapid screening and cost-effective infection detection.

Photo

News • mhealth

App accurately detects COVID-19 infection in people’s voices

Artificial intelligence (AI) can be used to detect COVID-19 infection in people’s voices by means of a mobile phone app, according to research to be presented at the European Respiratory Society…

Photo

News • Improved detection and treatment

New trial shows benefits of smartphone-based atrial fibrillation screening

Atrial fibrillation screening using conventional smartphones more than doubles the detection and treatment rate in older people compared to routine screening, new research finds.

Related products

Subscribe to Newsletter