NHS adopts AI imaging algorithm to improve COVID-19 treatment

Hospitals around the country will soon be able to use artificial intelligence (AI) to help diagnose COVID-19 patients and reduce the time needed to create treatment plans.

The 'COVID-19 Chest Imaging Database' (NCCID) has been pulled together by NHSX, the unit tasked with the organisation's digital transformation. The database includes over 40,000 CT scans, MRIs and X-rays taken during the course of the pandemic.

It is hoped that the database can speed up coronavirus diagnosis and help assess whether a patient will be likely to end up in a critical condition. Currently, the rapid rise in cases has put a strain on most of the country's hospitals, with reports that a person with COVID-19 symptoms is admitted every 30 seconds.

The health secretary, Matt Hancock, said that the NCCID was "a testament to how technology can help save lives in the UK".

An algorithm based on the NCCID images is currently being developed by clinicians at Addenbrooke's Hospital in Cambridge, which will help to inform more accurate patient diagnoses when they present to a hospital without a confirmed test.

Chest scans with visual signatures of the virus will be compared to patterns in previous scans held in the NCCID. With an understanding of the earlier stages of the disease, clinicians can determine appropriate medical intervention and reduce the potential for complications at a later stage.

"We are applying the power of artificial intelligence to quickly detect disease patterns and develop new treatments for patients," said Dominic Cushnan, the head of AI imaging at NHSX. "There is huge potential for patient care, whether through quicker analysis of chest images or better identification of abnormalities."

The NCCID is just one avenue of AI that NHSX is looking into. The NHS AI Lab has also launched a £140 million AI award for innovations that can bring benefits across the organisation.