Professor James Leiper
Associate Medical Director, British Heart Foundation
Artificial intelligence, once a far-off innovation, is becoming a reality in hospitals. Research we’ve funded is already driving AI development and enhancing diagnosis, treatment and care for heart patients.
MRI scans are crucial for diagnosing heart disease, but manual analysis takes, on average, over 13 minutes, consuming valuable time doctors could be spending with patients.
AI to accelerate diagnosis
With funding from the British Heart Foundation, scientists at University College London have developed new AI technology that can analyse scans and spot early signs of heart disease in just 20 seconds. The AI tool is currently being introduced at hospitals in the UK and around the globe.
The researchers trained the AI using heart scans from over 1,900 people. It measures the size of the left ventricle — the main pumping chamber of the heart — and how well it pumps blood around the body. The AI tool was found to detect changes to the heart’s structure and function with 40% greater precision and could identify information from scans that the human eye can’t detect.
The researchers estimate that their AI tool could save around 3,000 clinical days every year. By helping to accelerate diagnosis, it could also help heart patients get the treatment and care they need earlier on.
Women who have a heart attack 50% more likely
than men to receive the wrong initial diagnosis.
Closing the heart attack gender gap
The heart attack gender gap is costing lives, with women who have a heart attack 50% more likely than men to receive the wrong initial diagnosis. Clinical trials are testing an AI tool that could help close the gap.
University of Edinburgh researchers, funded by the British Heart Foundation, are developing an AI tool to diagnose heart attacks more quickly and accurately. Trained on data from over 10,000 people, almost half of whom were women, the AI combines patient information with blood test results to detect troponin — a protein released during a heart attack.
The tool was able to rule out a heart attack with 99.5% certainty, confirming patients were safe to go home, as well as identifying those who would benefit from further testing in hospitals. The tool diagnosed heart attacks with greater accuracy than current tests, especially in women.
New technologies have the potential to save many thousands of lives worldwide, and none of these advances would be possible without donations from our generous supporters. With ongoing support, we’ll continue to fuel research and drive forward new advances.