New AI test identifies COVID-19 within ONE HOUR

New AI test identifies COVID-19 within ONE HOUR of a patient’s arrival in A&E by analysing data from blood tests and vital signs

  • The Curial AI tool was developed by researchers from the University of Oxford
  • It can predict COVID-19 status in ‘near real-time’ from routinely collected data
  • In contrast, present nasal/throat swab-based tests take around 12–48 hours
  • Tests showed that the AI can predict if one has COVID-19 correctly 90% of time

A test powered by artificial intelligence (AI) can identify COVID-19 within one hour of a patient arriving in A&E by analysing their vital signs and blood test data.

Developed by researchers from University of Oxford, the system — dubbed the Curial AI — can make its diagnoses based on routinely gathered information alone.

At present, testing for the coronavirus involves the molecular analysis of special nose and throat swabs — with results typically taking around 12–48 hours.

A test powered by artificial intelligence (AI) can identify COVID-19 within one hour of a patient arriving in A&E by analysing their vital signs and blood test data (file photo)

According to the researchers, the tool can deliver near-real-time predictions of a patient’s COVID-19 status. 

In a study that ran from April 20–6 May, the team tested the AI tool on data from the 3,326 people who visited A&E at the four Oxford University Hospitals.

Curial had previously been trained using rich clinical data on 115,394 visits to A&E, along with 72,310 hospital admissions, comparing known cases of COVID-19 with control examples from before the pandemic to distinguish the differences.

Lead researcher and clinician Andrew Soltan said that the tool had accurately predicted a patient’s COVID-19 status in more than 92 per cent of cases — adding that the tool could prove useful for the NHS in the fight to stop the pandemic.

‘Until we have confirmation that patients are negative, we must take additional precautions for patients with coronavirus symptoms, which are very common,’ Dr Soltan added.

‘The Curial AI is optimised to quickly give negative results with high confidence, safely excluding COVID-19 at the front door and maintaining flow through the hospital.’

‘When we tested the Curial AI on data for all patients coming to Oxford University Hospitals’ emergency departments in the last week of April and the first week of May, it correctly predicted patients’ COVID status more than 90 per cent of the time.’

With their initial study complete, the researchers are now looking to carry out real-world trials of the software.

Developed by researchers from University of Oxford, the system — dubbed the Curial AI — can make its diagnoses based on routinely gathered information alone. At present, testing for the coronavirus involves the molecular analysis of special nose and throat swabs — with results typically taking around 12–48 hours

‘The next steps are to deploy our AI in to the clinical workflow and assess its role in practice,’ Dr Soltan continued.

‘A strength of our AI is that it fits within the existing clinical care pathway and works with existing lab equipment.’

‘This means scaling it up may be relatively fast and cheap,’ he explained.

‘I hope that our AI may help keep patients and staff safer while waiting for results of the swab test.’

A pre-print of the researchers’ article, which has not yet been peer-reviewed, can be read on the medRxiv repository. 

Source: Read Full Article