Coronavirus: Brighton researcher helps develop new test

University of Brighton senior lecturer Dr Louise Mackenzie was part of a crack UK team which thought up the innovative method, which they state is quicker than current tests.

Using blood and testing data from a Brazilian hospital, researchers found coronavirus patients tended to have decreased levels of certain white blood cells.

They also found Covid-19 cases generally had fewer platelets - fragments of blood cells which form clots and scabs.

So by drawing up a simple algorithm, Dr Mackenzie and the team were able to accurately predict whether a suspected patient was likely to have coronavirus based off a quick finger prick blood test.

The new test is able to predict whether someone is likely to have coronavirus so testing resources can be allocated efficiently

“In the UK we managed to reach 100,000 tests a day, but in Brazil the capacity is much smaller,” the biologist said.

“Hospitals were overrun with people who needed to be tested. That’s not fair. So we wanted to make something quick and cheap with stuff that’s already available.

“Here you could get tested in one hour if you’re rich enough.

“But that’s not applicable to places which don’t have that kind of money.”

The algorithm was able to predict whether a hospital patient had coronavirus with 93 to 94 per cent accuracy.

When testing people in the community, the accuracy was 80 to 84 per cent.

“That’s not bad for a small group of 30 to 40 people, but not perfect,” Dr Mackenzie said.

But the test would not be used to diagnose patients with coronavirus. Instead, it would help hospitals with limited resources narrow down the number of people who need testing.

Researchers are collaborating with medics in Gabon to use the test to combat future outbreaks

“If you have 10,000 people who need to be tested, this would help use your resources efficiently,” Dr Mackenzie said.

“It wouldn’t be for vulnerable people who need to get tested regardless, such as cancer patients.”

The test is still a work in progress. Researchers want to gain access to more data to improve the algorithm’s accuracy, a process called machine learning.

But there is already significant interest in the project.

The team is now working with a university in Gabon to help tackle future Covid-19 waves in the African nation.

• The coronavirus Sussex Crisis Fund has been set up to help those affected by the pandemic. The Argus’s charity and American Express have each donated £50,000 to kick-start the appeal. Grants will usually be for up to £5,000. More information is available at www.sussexgiving. org.uk/apply. To donate visit www.totalgiving.co.uk/appeal/sussexcrisisfund