Hospitals in Scotland are testing an artificial intelligence (AI)-based software that can detect heart attacks with high accuracy. The tool, called CoDE-ACS (Collaboration for the Diagnosis and Evaluation of Acute Coronary Syndrome), was developed with major contributions from the British Heart Foundation. The aim of the project is to improve the speed and quality of medical care and reduce mortality from myocardial infarction.
Difficulty in diagnosing heart attacks
Heart attacks are difficult to diagnose because symptoms such as chest pain, dizziness, or shortness of breath are common to many other diseases. Patients and even doctors can often attribute the symptoms to something else, leading to precious time lost. Without timely treatment, the probability of rapid death from this disease is about 70%.
Training CoDE-ACS for high accuracy
CoDE-ACS was trained on data from more than 10,000 patients hospitalized in Scotland with suspected heart attacks. According to the researchers, the tool is able to detect myocardial infarction with a probability of 99.6%. The tool analyzes the patient’s age, gender, medical history, examines electrocardiogram data, and uses a blood test to find troponin, a protein produced when the heart muscle is damaged. The tool expresses its analysis in a score from 0 to 100, with a higher score indicating a higher chance of a heart attack.
Saving lives through early diagnosis and treatment
“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 University of Edinburgh’s Center for Cardiovascular Science, who led the study published in the journal Nature. The researchers believe that this algorithm will allow doctors to identify people who come to the emergency room with a real life-threatening heart attack faster. This will greatly save time in treatment, as well as help to weed out patients with low risks to health and life in order to devote time to those patients who cannot wait.