An algorithm developed using artificial intelligence (AI) showed impressive speed and accuracy in determining the probability of heart attack during tests on thousands of patients, researchers reported.
Current guidelines recommend diagnosing myocardial infarction by measuring cardiac troponin levels against fixed thresholds. However, abnormal troponin levels are influenced by age, sex, comorbidities, and time from the onset of symptoms.
A study, led by the University of Edinburgh, assessed the effectiveness of an algorithm named Collaboration for the Diagnosis and Evaluation of Acute Coronary Syndrome (CoDE-ACS) to refine diagnosis. The algorithm was trained using data from 10,038 patients in Scotland, with a median age of 70 years, who had presented in hospital with possible myocardial infarction. Its performance was then tested using data from 10,286 patients (median age 60) with suspected myocardial infarction from seven cohorts in six countries.
The machine learning model integrated troponin levels alongside routinely collected patient information, including age, sex, ECG findings, and medical history, to score the probability of myocardial infarction on a scale of 0 to 100 for each patient.
Results of the study, published in the journal Nature Medicine, suggested that CoDE-ACS provided "excellent discrimination" by identifying more than twice as many patients as having a low probability of myocardial infarction in the external validation cohorts, with an accuracy of 99.6% (95% CI 99.4 to 99.7), compared with current diagnosis methods. Sensitivity from using the algorithm was 97.9% (95% CI 97.6 to 98.2).
Nicholas Mills, professor of cardiology at Edinburgh's Centre for Cardiovascular Science, 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."
Clinical trials of CoDE-ACS were underway in Scotland, aided by Wellcome Leap, which supports unconventional innovations, the University of Edinburgh said.
Professor Sir Nilesh Samani, medical director of the British Heart Foundation, which part-funded the research, said the AI system "could be transformational for emergency departments, shortening the time needed to make a diagnosis, and much better for patients".
Commenting on the study for the Science Media Centre, Steve Goodacre, professor of emergency medicine at the University of Sheffield, said: "This intriguing study shows how AI can use complex analysis, rather than a simple rule, to improve diagnosis. This doesn't yet show that we can replace doctors with computers. Experienced clinicians know that diagnosis is a complex business. Indeed, the 'ground truth' used to judge whether the AI algorithm was accurate was a judgement made by clinicians.
"It will be interesting to see how clinicians in the emergency department use this algorithm. What will they do if they think the algorithm has got it wrong? The next stage of the research will hopefully answer that question."
The research was funded with support from the National Institute for Health Research and NHSX, the British Heart Foundation, and Wellcome Leap.