AI-powered scans can identify people at risk of a fatal heart attack almost a DECADE in advance ‘by looking at the entire iceberg and not just the tip’
- The AI predicted heart risk with 90% accuracy, according to data
- Current medical scans are only able to see ‘the tip of the iceberg’
- It could benefit around 350,000 in Britain, cardiologists believe
- Government funding will fast track the tech into the NHS in two years
A high-tech heart scan developed by Oxford scientists can spot the risk of a heart attack nearly a decade in advance.
The new technology – which has received Government funding to fast track it into the NHS within two years – predicts heart risk with 90 per cent accuracy, according to data presented at the European Society of Cardiology congress in Paris.
The breakthrough could save thousands of lives a year by spotting at-risk patients whose heart problems cannot be detected using current scanning methods.
Cardiologists believe 350,000 people in Britain could benefit from the scans.
A high-tech heart scan developed by Oxford scientists can spot the risk of a heart attack nearly a decade in advance, which could save thousands of lives (stock photo)
Currently, patients experiencing chest pain are sent for CT scans.
Around a quarter have blockages which are treated with surgery, but the rest are sent home because the scans show no sign of narrowing of blood vessels which requires surgery.
But around one in five of these patients will go on to suffer a heart attack.
The new technology uses artificial intelligence to analyse routine CT scans to assess future risk.
It assesses build-up of fat and scarring around the heart – that would not usually be considered dangerous – and calculates whether this will cause problems in the future.
Trial results show it is 85 to 90 per cent accurate at predicting a heart attack in the next decade.
Professor Charalambos Antoniades, who led the study at Oxford’s Division of Cardiovascular Medicine, said: ‘What we can see with CT is just the tip of the iceberg.
‘We have developed an extremely powerful artificial intelligence system which looks at the entire iceberg – we are going from seeing just the tip to the whole, what is under the surface, which is the big part of the risk.’
He said the development was ‘massively important’ because patients found to be at-risk could be put on medication to improve their heart health.
‘It will direct treatment and it will save lives,’ he added.
New guidelines are set to recommend the number of people given CT scans for chest pain each year increases from 40,000 to 350,000.
Professor Antoniades said if the AI tool was used to analyse all of these scans which appeared normal it would save ‘tens of thousands’ of lives each year.
Professor Metin Avkiran, associate medical director at the British Heart Foundation, which funded the research, described the development as ‘a significant advance’.
‘This research is a powerful example of how innovative use of machine learning technology has the potential to revolutionise how we identify people at risk of a heart attack and prevent them from happening.
‘Such AI-based technology to predict an impending heart attack with greater precision could represent a big step forward in personalised care for people with suspected coronary artery disease.’