New AI Tool Predicts Heart Attacks Years in Advance

New AI Tool Predicts Heart Attacks Years in Advance

Due to the limitations of the currently most utilized diagnostic method, many patients are at risk of experiencing a heart attack or cardiac arrest without being identified as high-risk. British researchers have created a brand-new, artificial intelligence-powered solution to address these diagnostic flaws with the aim of saving lives in response to this reality. According to them, the tool is capable of detecting early signs up to 10 years before a heart attack occurs.

Myocardial infarction, commonly known as a heart attack, is a major cause of mortality worldwide. In the world, hundreds of thousands of individuals experiencing chest pain undergo a cardiac CT scan to identify constrictions or blockages in the coronary arteries, common sources of heart attacks. This examination, called “computed tomography,” or CT, serves as a standard diagnostic method, utilizing a combination of X-ray imaging and computer technologies to produce detailed images of the internal body structure.

However, a recent study by a group from the University of Oxford under the direction of Professor Charalambos Antoniades found that this technology is not entirely effective. To overcome the limitations of CT, scientists have developed an AI-powered tool. The system they have devised enables doctors to more accurately diagnose the risk of heart attacks. The tool was presented at the scientific sessions of the American Heart Association in Philadelphia, a conference where researchers share the latest advances in the field of cardiology.

The Limits of CT Scan

According to the study authors, in approximately 75% of cases, CT scans do not reveal any clear signs of artery narrowing. Consequently, many patients are falsely reassured after the examination, as they are considered healthy. However, some of them may occasionally die from a sudden heart attack. The reason is that certain vessel narrowings are undetectable via CT scans. In other words, this method is not sufficiently effective in detecting these more subtle and potentially fatal risks.

According to the team’s study involving 40,000 individuals who underwent CT scans, a significant number were diagnosed as “at risk.” However, an even larger number of patients who showed no signs experienced a heart attack in the following years.

Our study found that some patients presenting in hospital with chest pain – who are often reassured and sent back home – are at high risk of having a heart attack in the next decade, even in the absence of any sign of disease in their heart arteries,” explained Professor Antoniades, according to the university’s website.

The researchers’ AI tool is based on the analysis of two main data sets and a few clinical factors. It has been trained to detect narrowings of coronary arteries, including the smaller ones. It also examines changes in fat deposits around the arteries in cases of inflammation. Relative modifications can indicate an increased risk of a heart attack.

The goal is to assist doctors in refining their diagnostic methods to better identify at-risk patients. Equipped with this tool, they would be able to make more informed decisions about treatments to prevent heart attacks. The AI could detect early signs up to 10 years before a heart attack occurs.

During a trial of the tool, doctors were able to improve the treatment of 45% of patients who underwent routine CT scans in various British hospitals. Furthermore, the researchers suggest that their technology could decrease heart attacks by more than 20% and reduce cardiac deaths by 8% in patients who have undergone CT scans.