Patients could be set to benefit from more efficient and effective heart attack diagnoses thanks to a artificial intelligence-powered algorithm, new research has shown.
Researchers evaluated the tool, known as CoDE-ACS, by testing it on more than 10,000 patients across six countries around the world. They discovered that the AI could eliminate the possibility of a heart attack in more than double the number of patients compared to standard testing methods – all with an impressive accuracy of 99.6%.
With the researchers hoping streamlining diagnoses like this can help cut hospital admissions, clinical trials have begun in Scotland to evaluate whether the algorithm can alleviate pressure on emergency departments.
The best practice for diagnosing heart attacks is currently measuring a patient’s troponin levels, but with CoDE-ACS, clinicians can identify whether a person’s irregular troponin levels were down to a heart attack, rather than another condition.
However the issue with current best practice is that the same threshold is used for every patient, even though factors such as age, sex and other health problems affect troponin levels, meaning diagnoses can be inaccurate.
This ultimately leads to inequalities according to the British Heart Foundation whose previous research has shown that women are 50% more like to get an incorrect first diagnosis. Those who are initially misdiagnosed are at a 70% higher risk of dying after 30 days.
The AI tool could help stop these inequalities as it performed well regardless of age, sex and additional health complications.
The research was led by the University of Edinburgh and funded by the British Heart Foundation and the National Institute for Health and Care Research.
The British Heart Foundation’s medical director, Professor Sir Nilesh Samani, said: “Chest pain is one of the most common reasons that people present to Emergency Departments. 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.
“CoDE-ACS, developed using cutting edge data science and AI, has the potential to rule-in or rule-out a heart attack more accurately than current approaches. It could be transformational for Emergency Departments, shortening the time needed to make a diagnosis, and much better for patients.”
The tool was developed using data from 10,038 patients in Scotland who had presented at a hospital with a suspected heart attack.
The AI uses patient information – like age, sex, ECG findings, medical history and troponin levels – to assess the likelihood a patient has had a heart attack, eventually giving a probability score of 0-100.
Professor of cardiology at the University of Edinburgh, Nicholas Mills, who also led the research, added: “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.”
To access the full research report, click here.