The accepted standard for diagnosing myocardial infarction, or heart attack, involves assessing the blood for troponin levels. However, this approach applies the same benchmark for all patients, failing to take into account variables like age, gender, and pre-existing health conditions which can influence troponin levels, thereby potentially compromising the accuracy of diagnosis and leading to disparities. Now, an artificial intelligence (AI)-based algorithm offers a speedy way to exclude heart attack possibilities in patients and assists clinicians in discerning if irregular troponin levels are the result of a heart attack or a different condition. The AI tool functions efficiently regardless of the patient's age, gender, or other health conditions, demonstrating its potential in mitigating diagnostic inaccuracies and disparities across various demographics.

The AI algorithm, termed CoDE-ACS, was created utilizing data from 10,038 patients in Scotland who presented to the hospital with suspected heart attack symptoms. The algorithm uses routinely gathered patient data, such as age, gender, ECG results, medical history, and troponin levels, to estimate the likelihood of a patient having experienced a heart attack. The outcome is a probability score ranging from 0 to 100 for each patient. CoDE-ACS could potentially enhance the efficiency and effectiveness of emergency care by swiftly identifying patients who can safely be discharged, while simultaneously flagging those who require further hospital testing.

Researchers from the University of Edinburgh (Scotland, UK) evaluated the efficacy of the algorithm, termed CoDE-ACS, on 10,286 patients across six nations. Their findings revealed that CoDE-ACS was able to exclude the possibility of heart attacks in over twice the number of patients compared to traditional testing methods, with a remarkable accuracy rate of 99.6%. This capability of ruling out heart attacks more swiftly could substantially decrease hospital admissions. In addition to promptly excluding heart attack possibilities, CoDE-ACS could also support clinicians in identifying patients whose abnormal troponin levels are attributable to a heart attack rather than a different medical condition.

“For patients with acute chest pain due to a heart attack, early diagnosis and treatment saves lives,” said Professor Nicholas Mills, BHF Professor of Cardiology at the Centre for Cardiovascular Science, University of Edinburgh, who led the research. “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.” 

http://www.gzjiayumed.com/en/index.asp