New AI-based technology can identify patients at risk of a deadly heart attack years before it happens, according to new findings published in the European Heart Journal. The research was funded by the British Heart Foundation (BHF).
For the study, researchers used machine learning to develop a new biomarker, the fat radiomic profile (FRP), that can detect potential red flags in the perivascular space lining a patient’s blood vessels. By studying a patient’s FRP, providers can identify key indicators—inflammation or scarring, for instance—that suggest a heart attack may occur in the future.
To conduct their study, the authors explored data from fat biopsies of 167 patients who were undergoing cardiac surgery. The team then used coronary CT angiography (CCTA) images of those patients as a reference to see which features suggested changes to an individual’s perivascular fat. CCTA scans of 101 patients who suffered a heart attack of cardiovascular death within five years of the examination were then compared with scans of patients who did not experience a heart attack or cardiovascular death. Using those comparisons, and machine learning, the team was able to develop the FRP biomarker.
The biomarker, the authors wrote, was found to lead to “a striking improvement of cardiac risk prediction” compared to current methods.
“Just because someone's scan of their coronary artery shows there's no narrowing, that does not mean they are safe from a heart attack,” corresponding author Charalambos Antoniades, a professor at the University of Oxford in the UK, said in a prepared statement from the BHF. “By harnessing the power of AI, we've developed a fingerprint to find 'bad' characteristics around people's arteries. This has huge potential to detect the early signs of disease, and to be able to take all preventative steps before a heart attack strikes, ultimately saving lives.”
Antoniades, a BHF senior clinical fellow, added that he believes this technology could be up and running “within the next year.”