Somerset NHS Foundation Trust has used a new AI algorithm to detect signs of lung cancer in x-rays.
He conducted a trial using Behold.AI’s red dot algorithm which provided evidence that it could cut the time between initial X-ray screening and a CT scan in half.
The company said that over three months, the trust found that out of 3,794 images reviewed by the algorithm, the average scan-to-scan time was reduced from seven days to 2.8 days and results were delivered. to hospital systems in just 16 seconds.
The algorithm classified 562 cases as normal with a high level of confidence, of which radiologists disagreed with only 13. None of the discrepancies were considered clinically significant.
This could help the trust meet its 28-day cancer diagnosis target.
Buzz but little experience
Dr Paul Burn, consultant radiologist at the trust, said: “There has been a lot of buzz about AI at radiology meetings, but there is little experience of its use in an NHS trust. We embarked on a bottom-up initiative to test the algorithm, with the goal of helping us improve our SEO times.
“We have a fairly elderly patient population, which may make it more difficult for AI imaging solutions to be effective due to a higher incidence of abnormalities that show up on x-rays, such as scars and bumps. calcification.”
He added: “Normal high-trust results are a clear opportunity to learn where AI can be used in the future, especially for trusts with a large backlog reporting problem.”
The red dot algorithm was developed in conjunction with NHS consultant radiologists and provides two outputs: a subset of abnormal X-rays with a high probability of lung cancer and another subset of normal X-rays with a high confidence with a very high probability of being normal. .
Conquer the disease
Simon Rasalingham, CEO and President of Behold.AI, said: “Early-stage lung cancers are often missed by X-rays. We believe our technology can detect 22,000 additional lung cancer cases each year, which which gives these people a much better chance of beating the disease.
The trial was funded by Somerset, Wiltshire Avon and Gloucestershire Cancer Alliance (SWAG).