A new artificial intelligence (AI) algorithm developed by the Iscte - Instituto Universitário de Lisboa can predict the likelihood of a patient experiencing an acute myocardial infarction up to 13 minutes in advance, enabling proactive medical intervention.
The system was created in collaboration with the Faculty of Medicine at the University of Lisbon, which provided clinical data for training and testing the AI models. It was tested in real conditions at the Hospital de Santa Maria in cardiology intensive care.
According to Miguel Sales Dias, the principal investigator of the AIM Health project, the algorithm significantly anticipates the risk of ischemia, a condition where blood flow to the heart muscle is reduced. The system continuously monitors patients' physiological data and alerts medical staff when it detects relevant changes.
“We have developed a pilot that showed good results, and we are now working to implement it,” said Dias, adding that the goal is to expand the system to other cardiology intensive care units.
This AI application marks a shift from reactive to proactive healthcare, allowing for early therapeutic intervention before a heart attack occurs. “We can start treatment or correct damage before the patient suffers harm,” emphasized cardiologist Luís Rosário.
The AIM Health project has also developed a second advanced tool that uses computer vision and AI to assist cardiologists in diagnosing aortic valve calcification. This system automatically identifies the valve in MRI and echocardiography exams, distinguishing between calcified and healthy valves while quantifying the calcium present.
“The set of images used to train our algorithm would take a clinician 22 hours to analyze manually, but we do it in real-time,” noted Dias. Rosário added that the speed and accuracy of this technology facilitate early diagnosis, essential for personalized treatment planning, allowing doctors to focus on more complex clinical tasks.