Robot heart scanner could halve waiting lists
Doctors hopeful of cutting cardiac treatment backlog using AI system that gives results in 15 minutes
AN ARTIFICIAL intelligence heart scanner promises to cut the NHS backlog in half by delivering results within 15 minutes, scientists have revealed.
Created at Oxford University, the machine taught itself automatically to diagnose heart disease from images that would normally have been pored over by a team of doctors.
Crucially, the new system does not require the injection of a contrast dye. Patients currently have to lie in an MRI scanner for 20 minutes during which images are taken, then come out to be injected with the dye, then go back in for another 20 minutes.
The procedure is expensive and precludes some patients who cannot tolerate the dye or who are afraid of needles.
Known as Virtual Native Enhancement (VNE), patients using the new system enter the scanner for up to 15 minutes, with the results computed immediately. In a recent trial, VNE was found to detect signs of heart disease as accurately as four doctors using the standard technology.
It could be deployed for a wide variety of cardiac conditions, from hypertrophic cardiomyopathy, which is mostly seen in the young, to common heart failure, where the organ loses the strength to pump blood properly.
As of May there were 255,641 MRI scans waiting to take place, although not all will have been heart-related.
The Covid backlog saw 242,181 people waiting for invasive heart procedures – the highest number for May on record. Last month Sajid Javid, the Health Secretary, described the waiting list as “shocking”.
Dr Sonya Babu-Narayan, associate medical director at the British Heart Foundation, the charity which funded the development, said: “This research could have huge medical benefits as the same heart diagnosis could be made faster and without an injection, giving heart patients an easier experience. “AI technologies like these could slash the time taken to perform heart investigations. This means more people could reach the front of the queue for these vital heart tests sooner.”
The VNE system was built using a “deep learning” model that learned to recognise heart disease by studying more than 4,000 real-life images.
The next stage will be to set up a large-scale clinical trial across multiple hospitals.
The Oxford team believes the benefits will be so obvious to NHS chiefs that the technology will be rolled out within two years. Dr Qiang Zhang, who led the development, said: “The technology provides the same level of accuracy as the current practice but it’s quicker, cleaner and more pleasant for the patient. It cuts the time taken by more than half.
“We think it’s hugely significant and can be applied to most of the main heart conditions.”
The number of patients on hospital waiting lists could rise to 13 million within months as a result of Covid, Mr Javid has warned. More than 5.3million are thought to be on waiting lists.
Officials are worried that some seven million people who would have been expected in normal times to seek medical treatment did not come forward during the pandemic. These include people who have heart disease.
Mr Javid has promised to be “creative” in attempting to clear the backlog, but has warned the task would take “considerable time”.