New artificial intelligence (AI) technology that can identify if people are at high risk of having a fatal heart attack before it occurs has been approved in the EU.

The technology, CaRi-Heart, was developed by Oxford University spinout company Caristo Diagnostics.

The work that eventually led to the creation of Caristo Diagnostics was initially carried out by British Heart Foundation (BHF) researchers at Oxford University.

The CaRi-Heart technology can detect ‘invisible’ risk in people with possible heart disease, through the use of routine heart scans already performed in clinical practice.

It performs a ‘deeper dive’ into coronary CT angiogram (CCTA) scans, using AI and deep-learning technology to produce a Fat Attenuation Index Score (FAI-Score) which can accurately measure inflammation of blood vessels in and around the heart.

In a BHF study involving around 4000 patients, who were followed up for nine years following their original CCTA scan, researchers found that people with an abnormal FAI were up to nine times more likely to die of a heart attack in the next nine years compared to those with normal FAI scans.

The study also showed that at least one third of patients who underwent a routine CCTA and were initially considered low risk had a much higher risk after CaRi-Heart was applied to their scan.

Now that it has received a CE mark, the technology can be used across the UK and Europe and potentially rolled out for use on the NHS, the BHF said in a statement.

“The beauty of our technology is that it will not only save countless lives, but it is incredibly simple. CaRi-Heart analysis can be undertaken on any CT heart scan, hospitals don’t need to change equipment and patients don’t need another test,” said Cheerag Shirodaria, chief executive officer and co-founder of Caristo Diagnostics, and former BHF researcher.

“Physicians simply need to send their patient’s CT heart scan and they will then receive the personalised FAI-Score and CaRi-Heart Risk to guide patient management. It fits perfectly with a physician’s workflow,” he added.