AI Covid test for hospital patients is faster AND more accurate than lateral flows, study claims…and it doesn’t require a Pap smear
- Study of test on 72,000 NHS patients found it ruled out Covid 99.7% of the time
- Oxford experts said it could be a new first line of defense against Covid in NHS
- Lateral flow currently used, but if negative should be referenced with PCR
A new Covid test that would be better than a lateral flow could become the new first line of defense against the virus in hospitals – and it doesn’t require a painful Pap smear.
Experts from the University of Oxford claim they have developed an AI-based test that can detect the virus and give a result within 10 minutes without a throat or nose sample.
All a patient needs is a routine blood test, which is already taken from virtually everyone who is hospitalized.
A large study of the test – known as CURIAL-Rapide – in 72,000 NHS patients found it correctly ruled out infection 99.7 percent of the time.
The researchers said this was one-fifth more accurate than lateral flow devices, which are currently the most widely used rapid test.
When the new test is performed at the bedside of patients, the diagnosis can be made in just 10 minutes – compared to the half hour it takes on average for a lateral flow.
In emergency departments, CURIAL-Rapide produced results in 45 minutes by the time a sample was collected and processed.
This was faster than the hour required for lateral currents, which, if negative, must be compared to a PCR.
Many large hospitals have lab facilities where they can process PCR testing — the gold standard of Covid testing — on site.
But CURIAL-Rapide could make a “particularly big difference” in small hospitals that don’t have that luxury and have to steer away for results, the Oxford experts said.
A new Covid test that would be better than a lateral flow could become the new first line of defense against the virus in hospitals. It uses artificial intelligence and all it takes is a routine blood test (stock image)
When the new test is performed at the bedside of patients, the diagnosis can be made in just 10 minutes – compared to the half hour it takes on average for a lateral flow. In emergency departments, CURIAL-Rapide produced results in 45 minutes by the time a sample was collected and processed. This was faster than the hour required for lateral flows, which, if negative, should be compared to a PCR
Principal investigator Dr. Andrew Soltan, a clinician and machine learning researcher at Oxford, said: ‘Many of our patients who come to the ER are among the most vulnerable.
“CURIAL-Rapide is exciting because it uses new near-patient testing to collect all the data needed for a bedside prediction in 10 minutes, reducing the time to transport samples to a lab.
‘As a result, infected patients are identified earlier, while patients admitted with other conditions can be quickly and safely transferred to wards where they are less likely to be exposed.
‘This technology can help hospitals work more smoothly and can make a big difference, especially for smaller hospitals where there is no laboratory.
‘CURIAL is an example of how the collaborative power of the NHS, bringing together universities and hospital groups across the country, is helping to build a science base for safe and responsible use of clinical AI.’
The AI test uses routine blood tests and vital signs such as heart rate and blood pressure to look for signs of viral infection.
The Oxford team is in the process of applying for funding from the Department of Health for a national rollout.
The device was tested at five hospitals in Oxford, Birmingham, Portsmouth and Bedfordshire between December and March.
dr. Alex Novak, chief adviser for emergency medicine research at the Oxford University Hospitals NHS Foundation Trust, said the results were “exciting”.
He added that it had “the power to quickly translate into tangible benefits for patient care and optimize service performance on the front lines.”
Professor David Clifton, co-author of the study, said: “The university has shown that it has risen to the challenge of the Covid pandemic for society, and this AI-driven tool being implemented for use in emergency departments is a good one.” example of how we and the NHS can work together to produce something of lasting value for patients.
“With the emphasis shifting to living with Covid every year, research collaborations that deliver tools like CURIAL are the future for our field.
“Our recently announced Pandemic Sciences Center will be an important means to support such collaborations in the future, with the aim of putting us in a state of advanced preparedness for the arrival of new diseases or variants.”
Daily Covid hospital admissions in Britain exceed 1,000 for the first time since February
More than 1,000 Covid patients were hospitalized in a single day in the UK last week for the first time since February, official data show as the number of daily cases in the country rose again and deaths fell.
In its usual daily update, the Department of Health and Social Care revealed that 1,019 people had been admitted to the UK with the virus on 25 August.
It is the first time there have been four-digit Covid recordings since Feb. 24, when the second wave was brought under control and the jab rollout just gained momentum.
The DOH update — which often includes backlog hospital data because of the way it’s recorded — showed there were a further 943 Covid admissions on Aug. 26 and 901 on Aug. 27, both week-over-week increases.
Meanwhile, the picture was mixed as UK infections rose 4 per cent in one week to 32,181 but UK cases fell again, this time by 10 per cent.
Infection peaks in Scotland – attributed to schools going back in mid-August – and Northern Ireland, where vaccination rates have been slightly lower than the rest of the UK, will play a role.
But there are fears England and Wales could see things pick up again when classes resume this week and next, sparking a renewed debate on child pricking.
The government’s Covid dashboard also shows that 50 deaths have been recorded in the past 24 hours, a drop of more than 70 percent. The unusually low toll is believed to be caused by a withdrawal delay over the holiday weekend.