AI algorithm could speed up diagnosis of COVID-19 patients

behold.ai has announced that its artificial intelligence-based red dot algorithm quickly identifies chest X-rays from COVID-19 patients as ‘abnormal’, potentially speeding up diagnosis.

The behold.ai algorithm has been developed using more than 30,000 example images, all of which have been reviewed and reported by consultant radiology clinicians in order to shape accurate decision making. The red dot prioritisation platform is capable of sorting images into normal and abnormal categories in less than 30 seconds post image acquisition.

Dr Tom Naunton Morgan, chief medical officer at behold.ai, said: “The majority of deaths from COVID-19 are owing to pneumonia in the lungs of vulnerable patients. Pneumonia is a potentially life-threatening condition caused by a number of pathogens including, directly or indirectly, COVID-19 infection. Our algorithm can detect abnormal chest X-rays including pneumonia almost instantly. Out of 28 X-rays reviewed from patients with COVID-19, we correctly identified 85% of them as ‘abnormal’ using red dot. As we evaluate further positive cases from across the world, including here in the UK, our results will be further validated. This will increase the utility of our ‘instant triage’ and potentially help reduce the burden on healthcare systems as more and more cases of pneumonia present and require rapid diagnosis.”

The data follows recent news that behold.ai’s algorithm has been cleared by the US Food and Drug Administration (FDA). Commercial rollout in the US is planned for later this year. 

Simon Rasalingham, chairman and chief executive at behold.ai, said: “Our technology can make a big difference to patient safety, and the delivery of care and cost-savings to health services. It is available here and now to help manage the increased burden that will fall on health systems like the NHS in the coming weeks."