First author Dr Gavin Huangfu, from UWA’s Medical School, Harry Perkins Institute of Medical Research and Fiona Stanley Hospital, said coronary artery calcium scoring had revolutionised the prevention of heart disease in those with no symptoms, but the way it was measured had inherent drawbacks. It does not account for the location of calcified plaques along the coronary arteries despite knowledge that disease near the start of the artery carries greater risk, Dr Huangfu said.

Also, highly calcified plaques are assigned paradoxically higher risk, although are otherwise considered stable plaques and actually confer lower risk of cardiac events such heart attack or sudden death due to heart attack. Solutions to these flaws require analysis of each individual plaque, which is not feasible for a human investigator, but readily attainable using artificial intelligence.

The team found the algorithm, known as CAC-DAD score, was capable of measuring the burden of coronary calcification as well as each lesion’s distance from the coronary artery origin and reclassifying highly dense plaque as low risk all done with a single click. We found that the CAC-DAD score is more precise and effective than the standard Agatston score at predicting the risk of cardiac events, particularly around the time of surgery, in vulnerable populations, Dr Huangfu said. We also found the combined use of Agatston score and CAC-DAD score together further improved risk prediction, opening many possibilities for its clinical use.

Source: https://www.uwa.edu.au/news/article/2025/july/researchers-develop-more-precise-new-ai-tool-to-predict-risk-of-heart-attack#:~:text=The%20team%20found%20the%20algorithm,done%20with%20a%20single%20click