Clinicians could soon use an algorithm developed using artificial intelligence (AI) to improve the speed and accuracy of heart attack diagnosis in emergency departments, new research has shown.
The ability to quickly rule out a heart attack dramatically reduces hospital admissions, and the new tool, which was found to have an accuracy of 99.6 per cent, could help ease pressure on emergency departments.
Heart attacks are currently diagnosed by measuring levels of the protein troponin in the blood, but the same threshold is used for every patient. Variables such as age, sex and other health problems are not considered, affecting the accuracy of the test and creating inequalities in diagnosis, the paper suggests.
The newly developed algorithm known as CoDE-ACS was created using data from patients in Scotland who arrived at hospital with a suspected heart attack. The median age of the patients was 70, and 48 per cent were women.
The tool was then tested on over 10 000 patients in six countries worldwide and uses patient information such as age, sex, ECG findings, medical history, and troponin levels to predict the probability that an individual has had a heart attack.
Compared to current testing methods, the new machine learning tool was able to rule out a heart attack in more than double the number of patients, with a very high level of accuracy across diverse populations, showing its potential to reduce misdiagnosis.
Previous research has shown that women are 50 per cent more likely to have an incorrect initial diagnosis, and misdiagnosis has severe consequences for patients, increasing the risk of death by 70 per cent within 30 days.
The AI tool was also useful for looking at troponin in a patient’s blood and allowed doctors to identify whether high troponin levels were due to a heart attack or related to another condition, researchers said.
Professor Nicholas Mills, from the University of Edinburgh, who led the study, 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 straightforward.’
Using the algorithm, emergency departments could rapidly identify patients who were safe to go home and those who needed to stay for further tests, improving the efficiency of emergency care.
Professor Sir Nilesh Samani, medical director at the British Heart Foundation, added: ‘CoDE-ACS 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.’
Clinical trials are underway in Scotland to assess the tool’s impact in accident and emergency departments.