Key points from article :
An AI system developed at Oxford University can predict heart attacks up to 10 years in advance by analyzing CT scans, potentially saving thousands of lives annually.
The technology, under review by the UK's National Institute for Health and Care Excellence, identifies hidden inflammatory damage in coronary arteries that traditional CT scans miss.
This AI system was trained on US case studies and validated with data from 40,000 UK hospital patients, demonstrating its effectiveness in identifying high-risk individuals.
In 45% of cases, doctors changed patient treatment plans based on the AI's risk assessment, including prescribing preventative medications like statins or colchicine.
Published in The Lancet, the research was led by Professor Charalambos Antoniades at Oxford University's Radcliffe Department of Medicine and funded by the British Heart Foundation.
The researchers are now developing similar AI systems to predict strokes and other conditions like diabetes, with plans to expand the heart attack technology's use in the US and Europe.
