A screening system for Coronavirus which analyses the personal and unique coughing sounds of a million people across the country could be on the horizon.
The Novoic COVID app works by recording someone’s cough and using an algorithm to analyse cough patterns to screen them for the virus.
Novoic is now asking for one million volunteers to ‘donate their cough’ so that they can differentiate reliably between the sound of COVID-19 infected patients and those without the virus.
It believes the app could be rolled out as early as August if they get enough data.
With a vaccine potentially a year away, many public health experts believe the best way of containing the virus is by testing, tracing contacts and isolating patients if necessary. This will require large scale screening.
Novoic is a health start-up founded by Oxford and Cambridge researchers, who say that current wet-lab-based tests can’t meet this need as they are relatively expensive, scarce and slow. Compared with an at-home test, in-person visits increase the risk to members of the public and health personnel.
Emil Fristed, co-founder and CEO of Novoic, said: “When the respiratory system is affected by a disease, it can change the sound of your breathing, coughing and vocal quality. Using sound analysis these changes can be measured. We already know that other respiratory conditions can be detected from cough patterns. COVID-19 affects the respiratory system in a unique way. So, it’s likely that the cough patterns are unique too.
“Different people’s voices of course sound different from each other, including when they’re healthy. To build accurate algorithms that work for everyone, we need a lot of data, which is why we are calling upon the public to step forward. If we capture enough cough sounds, we believe this could be the answer to cheap, accessible testing.
“The pandemic is impacting us all. We wanted to do what we could to help, with the one thing we do best: analysing vocal and language patterns to detect diseases.”
Novoic was founded at the beginning of 2019, is venture capital-backed, and is being accelerated at the Oxford Foundry, an entrepreneurship centre at the University of Oxford, through its L.E.V8 accelerator and COVID-19 Action Plan.
The company uses advanced speech analysis to detect brain diseases, such as Alzheimer’s and Parkinson’s. It is now using that same technology to fight COVID-19.
