Artificial intelligence is also revolutionizing cardiology, recognizing the risks that can lead to a heart attack within a decade.

Cardiovascular diseases are the most common cause of death worldwide, causing 17.3 million deaths annually. A number that could rise to 23.6 million by 2030.

At the same time, however, due to prevention and better medical care, mortality from heart attacks has decreased significantly. And it could be reduced even more with the help of artificial intelligence.

Artificial intelligence has the ability to detect potential risks in time. From the data of a single exam, AI can spot patterns or outliers much faster than even the most trained staff.

Recognizing the risk of heart attack

Artificial intelligence could, for example, save the lives of thousands of patients suffering from chest pain who, despite undergoing a CT scan, were not found to be at risk of a heart attack – which is because very small narrowings are often in the coronary vessels they are not perceived. If inflamed, they may rupture and block arteries, causing a heart attack. Until now, it was impossible to identify this risk in advance.

Oxford researchers have now developed an artificial intelligence tool that can detect the presence of such abnormalities even in patients who do not have significant coronary artery narrowing, thereby predicting their risk of heart attack over a decade. These are the findings of a study funded by the British Heart Foundation and presented at the American Heart Association’s scientific sessions in Philadelphia.

Artificial intelligence provides the solution – here too

In the UK around 350,000 people have a CT scan of the heart every year – the standard test to detect narrowing or blockages in the coronary arteries.

In the aforementioned study, the research team, led by Professor Charalambos Antoniadis from Oxford University’s Radcliffe Department of Medicine, analyzed the examinations of 40,000 people from eight British hospitals. In about 3/4 of the cases, no clear signs of significant strictures were detected, resulting in the patients being discharged immediately. The analysis of the data, however, showed that not only patients with significant narrowing of the coronary vessels are at risk.

Many patients who eventually had a heart attack or died of heart failure were not found to be at risk because the smallest signs were missed. But now they too can be detected: not by humans, but by artificial intelligence.

The research team’s AI tool looks at the fat around the inflamed arteries to calculate the true risk, as such variations indicate an increased risk of heart attack.

Targeted and customized treatment

In a pioneering pilot project the medical team developed an AI-assisted risk assessment for 744 patients. Based on this evaluation the responsible doctors decided to change the treatment in 45% of the cases. According to the study, the use of this technology reduced heart attacks by more than 20% and deaths and strokes by 8% among the people tested.

“Many people die needlessly from heart attacks every year. It is of crucial importance to utilize artificial intelligence to shape the treatment of patients”, says Charalambos Antoniadis. “Here we have shown that the course of treatment can be changed or even improved if doctors have a clearer picture of a patient’s risks.”

Saving time and money

This new tool is just one example of how artificial intelligence can contribute to the early initiation of treatments, thus saving lives, significant time, and a lot of money. Another example is the “Decision Support System” (Decision Support System) developed by British cardiologists in mid-2023, which could speed up the diagnosis of a heart attack in emergencies with the help of artificial intelligence.

The AI ​​tool can determine blood troponin value faster. This is the protein that plays a key role in the diagnosis of heart attacks. Until now patients are often admitted to the hospital or have to stay longer for observation until the troponin value is determined, even though they have not had a heart attack. Another case in which TN could save the patient time and money.

Early recognition of risks

Compared to humans, AI can identify risks not only faster, but sometimes more reliably. In mid-2023 American researchers presented an artificial intelligence tool that detects suspicious patterns in a cardiogram and detects dangerous blockages in the coronary arteries with greater reliability than an experienced cardiologist.

The earlier the risk is identified, the lower the risk to the patient. In a routine procedure, a so-called stent can be placed in the patient, long before the appearance of a dangerous blockage. A stent is a small, plastic or metal tube, which is placed in a coronary artery, thus preventing it from closing and thereby preventing the risk of a heart attack.