Coronavirus testing is finally available in the UK, but it’s still a long-winded process, often involving samples being sent to remote labs.
But a new project could offer an easier way, using an app to detect the telltale dry cough caused by COVID-19, the Oxford Mail reports.
Oxford-based start-up Novoic is asking volunteers to submit coughs via their website, and will build up a database of coughs from both healthy and coronavirus-infected patients.
The company is also developing a language-based tool to detect the early stages of neurological diseases.
The idea is to create an algorithm which can differentiate between “healthy” coughs and those caused by coronavirus, the Oxford Mail reported.
The company says that current tests are “expensive and slow” and also pose a risk as they require an in-person visit.
The company hopes to build a database of a million coughs by the end of July, and is relying on volunteers to deliver a cough via their website.
Volunteers will also be asked about whether they have been tested for COVID-19, the outcome of their test, and their symptoms, as well as some general questions.
The idea of using automatic speech recognition to detect illnesses is not knew, the company says.
The company says previous research has used automatic speech recognition and acoustic analysis to identify speakers suffering from a cold, and to provide information on individuals with asthma or cancer.
“The respiratory system is key for humans to produce voice – when air from the lungs passes through and is shaped by the airways, the mouth and nasal cavities,” it says.
“When the respiratory system is affected by a disease it can change the sound of you breathing, coughing, and the vocal quality.
“Specific cough characteristics have already been shown in respiratory diseases such as asthma, pneumonia, whooping cough – using advanced analysis,” Novoic claims.
All information collected is anonymous, and will be used to fine-tune an algorithm which the company hopes could radically speed up diagnosis.
“We aim to create an application that can reliably differentiate between the cough and voice of COVID-19 infected patients, and those with other respiratory conditions. With enough data collected we can make our machine learning algorithms smarter.”
Coronavirus: what happened today
Read more about COVID-19
How to get a coronavirus test if you have symptoms
How easing of lockdown rules affects you
In pictures: How UK school classrooms could look in new normal
How public transport could look after lockdown
How our public spaces will change in the future