New app detects Covid in your voice – and is 'more accurate than lateral flows' – The Mirror

A new app can detect coronavirus in under one minute by listening to your voice and proves more accurate than lateral flow tests – promising to revolutionise how the virus is dealt with
A new app can detect Covid in under one minute just by listening to your voice – and claims to be more accurate than lateral flow tests.
The scientific breakthrough is powered by artificial intelligence, is easier to use and more accurate than the commonly used LFTs as well, according to the app creators.
The mobile app gives an accurate positive result 89 per cent of the time and 83 per cent of the time for negative cases.
It delivers its answer in under a minute compared to the unwieldy LFTs where accuracy reportedly varied depending on brand.
The app could revolutionise the approach to testing for Covid especially in poorer countries where the gold-standard PCR tests are expensive and often difficult to distribute.
In the UK, it could help keep the virus at bay in the long-term after the government decided to scrap free LFTs.
The Dutch boffins behind the new app explained how it works.
They said that Covid usually affects the upper respiratory tract and vocal chords of a person, leading to subtle changes in their voice.
The team investigated whether using this to detect the novel virus was possible.
During development, they used data from the University of Cambridge’s crowdsourcing COVID-19 Sounds App.
This contained 893 audio samples from 4,352 participants, 308 of whom had tested positive for the virus.
The app would take some basic info on participants including demographics, medical history and smoking status.
It would then ask them to record some respiratory sounds which include coughing three times, breathing deeply, through their mouth and reading a short sentence three times.
Then a voice analysis technique called Mel-spectrogram analysis was used which identifies different key features in the voice – including loudness, power and variation over time.
Then, an AI-based on human neural networks was used to work with this to detect the deadly virus.
Researcher Wafaa Aljbawi, from the University of Maastricht, said: “These promising results suggest that simple voice recordings and fine-tuned AI algorithms can potentially achieve high precision in determining which patients have COVID-19 infection.
“Such tests can be provided at no cost and are simple to interpret. Moreover, they enable remote, virtual testing and have a turnaround time of less than a minute.
“These results show a significant improvement in the accuracy of diagnosing COVID-19 compared to state-of-the-art tests such as the lateral flow test.
“The lateral flow test has a sensitivity of only 56 per cent, but a higher specificity rate of 99.5 per cent.
“In other words, we could miss 11 out 100 cases who would go on to spread the infection, while the lateral flow test would miss 44 out of 100 cases.
“The high specificity of the lateral flow test means that only one in 100 people would be wrongly told they were COVID-19 positive when, in fact, they were not infected, while the LSTM test would wrongly diagnose 17 in 100 non-infected people as positive.
“However, since this test is virtually free, it is possible to invite people for PCR tests if the LSTM tests show they are positive.”
Before the app can begin to appear on phones everywhere the team say they need more participants.
Since the start of the project 53,449 audio samples from 36,116 participants have now been collected and can be used to improve and validate the accuracy of the model.
The findings will be presented at the European Respiratory Society International Congress in Barcelona, Spain.
Get our daily health briefing with all the news you need to know direct to your inbox

source