An artificial intelligence system that can identify people who are likely to have a heart attack up to 10 years in the future could soon be up and running across Britain.
The technology, which could save thousands of lives a year, is being evaluated by the National Institute of Hello and Care Excellence (Nice) and a decision on its use in the NHS is expected by the end of the year.
Scientists on the project also revealed that they are working on similar AI systems to predict whether someone is at risk of suffering a stroke and to detect those at risk of conditions such as diabetes.
“This technology has already been tested in several hospitals in the UK and the results are extremely encouraging,” said Prof Charalambos Antoniades, head of the Orfan (Oxford Risk Factors and Non-Invasive Imaging) study. “If installed nationally, it will help save thousands of people from early heart attacks or death from heart disease.”
More than 300,000 people in Britain each year suffered from severe chest pain and had a CT scan to determine whether they were suffering from heart abnormalities such as arterial blockages, Antoniades said. However, less than 20% of those scanned were found to have blockages or dangerous narrowing of their coronary arteries. “The remaining over 80% show no abnormalities. They are sedated and sent home, very often without any medication,” said Antoniades, chairman of cardiovascular medicine at Oxford University. “However, these assurances are often misplaced.”
In fact, about two-thirds of this “safe” group go on to suffer serious — sometimes fatal — heart events, including heart attacks. “It is clear that we have missed signals from our scans that could have told us about those who are in real danger,” he added. “This is a huge health problem, and we believe AI is the perfect technology to address it.”
The surveyled by a team from Oxford University’s Radcliffe Department of Medicine and published in Lancet last week, is designed to spot abnormalities that are missed on standard CT (computed tomography) scans. This knowledge would allow doctors to prescribe preventive treatments such as anti-inflammatory drugs to patients.
A big part of the problem was that damage to an artery caused by inflammation was not detected by a CT scan, Antoniades said. “Our discovery was to find a way to bring out hidden information by using AI to enhance our CT images to show what damage has been done. In the past we couldn’t imagine this, but now we can.”
The technique uses data on the characteristics of coronary plaques as well as changes in the fat around inflamed arteries to provide key information about the health of our heart arteries. “Essentially, these readings tell us what the patient’s absolute risk is of having a fatal cardiac event in the next 10 years,” he said.
These risk factors were originally developed through case studies in the US, but the data have since been evaluated using 40,000 patients in UK hospitals.
after the promotion of the newsletter
“What we found was clear. Patients who were shown to have severe inflammation in their coronary arteries were found to be at extremely high risk of severe heart disease, such as heart attack. We have found a way to identify the hidden factors that lead to a heart attack.
The study, funded by the British Heart Foundation, revealed that in 45% of cases, clinicians decided to change the patient’s treatment in view of the data provided by the AI analysis. These treatments include giving high doses of statins or drugs such as colchicine, which are known to reduce the risks of cardiovascular disease.
Antoniades added: “We also plan to expand the supply of this UK-made technology to the US, where it is also being evaluated by the Food and Drug Administration, and to Europe, where it is already approved for clinical use.”
