This article is more than 1 year old

AI to detect heart attacks tested in the land of the deep-fried Mars bar

Scotland's emergency rooms don't always have the resources to make good diagnoses under pressure

Hospitals in Scotland are trying out AI software that can determine whether a patient is suffering from a heart attack, in an effort to improve accident and emergency response times.

Heart attacks can be difficult to detect as its symptoms – including chest pain, dizziness, and shortness of breath – are associated with many other conditions. Physicians often miss other vital signs. If misdiagnosed and left untreated, a patient has "a 70 per cent higher risk of dying after 30 days," the British Heart Foundation (BHF) claimed.

The AI system being tested, meanwhile, can "rule out a heart attack in more than double the number of patients, with an accuracy of 99.6 percent," the not-for-profit org suggested.

"Chest pain is one of the most common reasons that people present to Emergency departments," Sir Nilesh Samani, medical director of the BHF, commented. "Every day, doctors around the world face the challenge of separating patients whose pain is due to a heart attack from those whose pain is due to something less serious."

The charity helped fund the research and development of CoDE-ACS (Collaboration for the Diagnosis and Evaluation of Acute Coronary Syndrome) tool to diagnose heart attacks. The system, powered by machine-learning algorithms, predicts the likelihood patients are suffering from a heart attack.

CoDE-ACS analyzes a patient's age, sex, medical history, inspects electrocardiogram data, and uses a blood test to look for troponin – a protein that appears when heart muscles are damaged – to calculate a score out of 100. Higher scores mean there is a higher chance of heart attack. The researchers believe that the algorithm can sort out whether some people checking into A&E with symptoms are really having heart attacks or not, allowing physicians to identify those more at risk more quickly.

"For patients with acute chest pain due to a heart attack, early diagnosis and treatment saves lives," explained Nicholas Mills, professor of Cardiology at the Centre for Cardiovascular Science at the University of Edinburgh, who led the research published in Nature.

"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."

CoDE-ACS was trained on data from over 10,000 patients who were admitted to a hospital in Scotland with a suspected heart attack.

The technology is being trialed, again in Scotland, to see if it can improve care in Accident and Emergency departments, the BHF said.

"If adopted in practice, the CoDE-ACS clinical decision support system could reduce time spent in emergency departments, prevent unnecessary hospital admission in patients unlikely to have myocardial infarction and at low risk of cardiac death, and improve the recognition and treatment of those with myocardial infarction rather than myocardial injury, with benefits for both patients and health care providers," the study concluded. ®

More about


Send us news

Other stories you might like