Scientists have created an artificial intelligence model that can detect if someone has coronavirus - purely from the sound of their cough.

The AI algorithm, built in the US, correctly identified patients who were infected even if they had no other symptoms.

Researchers say humans are unable to hear the vital difference in the sound of someone with a cough who is asymptomatic.

They say this is because Covid-19 changes the way you produce sound, even if there are no symptoms.

The model, developed by the Massachusetts Institute of Technology (MIT), was 98.5% successful in detecting people who had officially tested positive for Covid-19.

And it could tell an impressive 100% of cases of people who had no other symptoms.

AI expert Calum Chace said the model was "a classic piece of AI” which showed it could be helpful, the BBC reported.

He added: "It's the same principle as feeding a machine a lot of X-rays so it learns to detect cancer.

"It's an example of AI being helpful.

"And, for once, I don't see a lot of downside in this."

Researchers collected around 70,000 recordings of coughs, with 2,500 from confirmed Covid patients.

The MIT team says the tool could be used daily for students and workers as they return to schools and jobs.

This would help countries as they return to normal from lockdown restrictions.

It could also quickly identify outbreaks of the virus in communities.

To turn the algorithm into an app so people can use it on their mobile phones, the developers will need approval from the regulators.

Researchers wrote: “AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale Covid-19 asymptomatic screening tool to augment current approaches in containing the spread of Covid-19.

“Practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups.”

Cambridge University has been developing a similar project, which had an 80% success rate in identifying positive cases from breath and cough sounds.

The findings were published in the IEEE Journal of Engineering in Medicine and Biology.