AI tool predicts three recovery paths for heart attack survivors

Researchers from the University of Surrey have developed an AI tool that identifies three distinct health trajectories in heart attack survivors over five years, potentially enabling earlier, personalized care. The tool predicts which path a patient will follow at the time of the event itself. The study, published in the Journal of the American Medical Informatics Association, analyzed health records from 12,701 UK Biobank participants. It found 63 percent developed cardiometabolic conditions, 23 percent suffered lung and organ decline linked to smoking, and 14 percent developed structural heart diseases. The smoking-related group had a 44 percent mortality rate, over three times higher than the largest group. Genetic analysis confirmed each trajectory mapped to distinct molecular pathways, validating the findings as biologically meaningful. While traditional risk scores remain the strongest single mortality predictor, the AI tool adds crucial detail on why and where intervention is needed, though researchers emphasize the approach is still in early stages.