Internet Explorer is no longer supported by Microsoft. To browse the NIHR site please use a modern, secure browser like Google Chrome, Mozilla Firefox, or Microsoft Edge.

Artificial intelligence could speed up heart attack diagnosis

Published: 15 May 2023

New research funded by NIHR and the British Heart Foundation could soon be used by doctors to diagnose heart attacks more quickly and accurately. The research was led by scientists at the University of Edinburgh, who tested an algorithm on over 10,000 patients around the world.

The algorithm, called CoDE-ACS, uses patient information including age, sex, heart measurements and other test results. This is combined to give a score of how likely it is that each patient has had a heart attack. By quickly spotting patients who are very unlikely to have had a heart attack, the algorithm can ensure they’re sent home. And it can make sure patients who are at higher risk stay in hospital for further tests.

The algorithm was tested on patients in 6 countries around the world. Researchers found that CoDE-ACS was able to rule out a heart attack in more than double the number of patients, compared to current tests. It had an accuracy of 99.6 per cent.

Improving on current tests

At the moment, the best way to diagnose a heart attack is to measure levels of a protein called troponin in the patient’s blood. But the same threshold is used for every patient. This means that other factors that might affect troponin levels aren’t taken into account.

CoDE-ACS combines troponin measurements with other key information. This gives a more accurate estimate of whether a particular patient has had a heart attack.

The tool is now being tested in clinical trials in Scotland with support from Wellcome Leap, to see whether it can help reduce pressure on Emergency Departments. The research was funded by NIHR through the NHS AI Lab’s AI in Health and Care Award, and published in Nature Medicine.

Professor Nicholas Mills, BHF Professor of Cardiology at the Centre for Cardiovascular Science, University of Edinburgh, who led the research, said:

“For patients with acute chest pain due to a heart attack, early diagnosis and treatment saves lives. Unfortunately, many conditions cause these common symptoms, and the diagnosis is not always straight forward. Harnessing data and artificial intelligence to support clinical decisions has enormous potential to improve care for patients and efficiency in our busy Emergency Departments.”

Steve Barclay, the health and social care secretary, said: “Ground-breaking research like this — backed by over £134,000 of government funding through the AI in Health and Care Awards — has the potential to improve diagnosis and speed up access to treatment for a range of conditions like heart attacks.

“This tool this could also reduce pressure on busy A&E departments and support clinical decision-making. We continue to invest in the latest technology, including AI, to help us cut waiting lists so people can get the care they need more quickly.”

Latest news