University of Surrey researchers develop AI to predict heart attack recovery

Researchers at the University of Surrey have developed an artificial intelligence tool designed to predict recovery trajectories for patients following an acute myocardial infarction (heart attack). The study, published in the Journal of the American Medical Informatics Association, analyzed the longitudinal health records of 12,701 UK Biobank participants.

The AI identified three distinct five-year recovery patterns. The largest group, comprising 63% of participants, developed cardiometabolic conditions such as hypertension, type 2 diabetes, and dyslipidemia. A second group of 23%, likely consisting of smokers, experienced progressive declines in lung, musculoskeletal, and other organ systems. The final 14% developed structural heart disease, arrhythmias, and kidney disorders.

Notably, the smoking-related trajectory showed a 44% mortality rate, which is more than three times higher than the rate observed in the cardiometabolic group. Lead author Dr. Anthony Onoja stated that the tool can predict these trajectories at the time of the initial heart attack using demographic data and pre-existing diagnoses. This capability aims to allow clinicians to provide earlier, personalized, and risk-aligned follow-up care.