New AI tool helps personalize care after heart attacks

The health of heart attack survivors follows three distinct paths in the five years after their health event, say researchers from the University of Surrey. The research used an AI tool to find patterns that could not only help clinicians predict which patients could follow which path - but it could also allow them to tailor care sooner through their recovery.

The study, published in the Journal of the American Medical Informatics Association, used the health records of 12,701 UK Biobank participants who had experienced a heart attack. The team tracked the sequence and timing of new diagnoses after the event, then used data-driven machine learning to group patients who followed similar patterns.

The Surrey team found that 63 per cent developed cardiometabolic conditions such as hypertension, type 2 diabetes and dyslipidemia, alongside episodic heart and respiratory complications. The team also found that 23 per cent, who were thought to be smokers, suffered a decline in the lungs, musculoskeletal system and other organs. Around 14 per cent of patients developed structural heart diseases, arrhythmias and kidney problems.

The smoking-related group had a mortality rate (44 per cent) that was more than three times the rate of the largest group (with cardiometabolic conditions).

Dr Anthony Onoja, lead author of the study and Research Fellow from the University of Surrey, said:

"We found that we could predict the health trajectory a patient would follow after a heart attack, at the point of the event itself, using their pre-existing diagnoses and demographic data. Our AI tool was incredibly effective at finding and predicting the highest-risk group, where respiratory conditions, older age, and higher deprivation scores were key predictors.

"Our approach is exciting, but we are still early in this journey, and we believe that in the future this could help hospitals identify people who follow these trajectories early and develop tailored care for them."

The research team also investigated whether the patterns were biologically meaningful, and genetic analysis confirmed that each group mapped to distinct molecular pathways - immune activation and tissue remodelling in the largest group, insulin signalling and lipid transport in the arrhythmia group and chronic inflammation and degeneration in the smoking-related group.

Professor Nophar Geifman, senior author of the study from the University of Surrey, said:

"Clinicians typically use risk assessments, such as the SMART score, to help them understand how likely a patient is to have another heart event. We found that these tools are still the strongest single predictor of mortality in our study, but the trajectories added detail that a stand-alone score cannot provide. The patterns we have identified show that we can capture more than just a patient's risk but, crucially, why, and where intervention could be needed."