Scientists in the UK have found a way to identify people most likely to have a heart attack years before it strikes.
he breakthrough by Oxford University uses artificial intelligence to look "beneath the surface" of routine CT scans and spot changes to blood vessels supplying the heart.
Researchers said 350,000 people a year could benefit from the checks - found to be up to 90pc accurate.
And they could be rolled out across the NHS in as little as two years - ensuring high-risk patients get the right treatment to avert potential problems. The scans will typically be offered to people aged 40 to 70 with chest pains or who are considered at particular risk of heart attacks because of obesity, smoking or diabetes.
Currently, patients experiencing chest pains are sent for CT scans which show blockages in about 25pc of cases.
While some patients are offered surgery, most are sent home without treatment, despite the fact many will later go on to have a heart attack.
CT scans are currently used on about 40,000 high-risk patients each year.
The new technology can detect a dangerous build-up of fat and scarring around the organ up to nine years before they reach dangerous levels.