Can AI Predict Heart Attacks and Strokes?

What if AI could predict your risk of a heart attack or stroke? Researchers are speeding up the analysis of common imaging tests to quickly identify people at risk.

A heart attack strikes suddenly and often without warning. Unless treatment is started quickly, the heart muscle begins to die. What if there were a way to predict your risk of a heart attack in the future, so you could take steps to prevent it?

An international team led by researchers at Cedars-Sinai Medical Center in Los Angeles has developed an artificial intelligence (AI) tool, which they say can quickly and accurately predict your risk of a heart attack in five years based on the amount of plaque buildup in arteries supplying your heart. The researchers reported their findings in The Lancet Digital Health.

YOU MIGHT ALSO LIKE: 4 Surprising Heart Attack Risk Factors

Using AI to measure plaque

The study used machine learning, a type of AI that trains machines to search imaging scans for patterns associated with conditions like heart disease.

In the study, investigators used an imaging test called coronary computed tomography angiography (CCTA) to measure plaque in coronary arteries. Plaque buildup from coronary artery disease can narrow these blood vessels to the point where less blood reaches the heart. Eventually, the loss of blood flow can lead to a heart attack.

Doctors often use CCTA to check for narrowed or blocked arteries and diagnose coronary artery disease. But there is no quick and easy way for them to measure the plaque they find. It's a manual process that takes a trained professional 25 to 30 minutes to complete per scan.

The investigators trained an AI algorithm to identify and measure plaque deposits on CCTA scans in around five seconds. The AI measurements matched those of expert readers, as well as results of two invasive tests (intravascular ultrasound and catheter-based coronary angiography) that are considered very accurate at analyzing plaque buildup. AI measurements also accurately predicted heart attack risk within 5 years in people who were part of the Scottish Computed Tomography of the Heart (SCOT-HEART) trial.

The study's senior author, Damini Dey, PhD, said the results are still preliminary but promising.

"More studies are needed, but it's possible we may be able to predict if and how soon a person is likely to have a heart attack based on the amount and composition of the plaque imaged with this standard test," according to Dey, director of the quantitative image analysis lab in the Biomedical Imaging Research Institute at Cedars-Sinai.

AI predicts heart disease risk with x-ray

In another study, presented at the annual Radiological Society of North America meeting, investigators reported that they had "trained" a deep learning model to predict the 10-year risk of dying from a heart attack or stroke using a single chest x-ray.

The model they developed, called CXR-CVD risk, might be used to determine which patients could benefit from cholesterol-lowering statin medications. Today, doctors use the atherosclerotic cardiovascular disease (ASCVD) score, which includes factors like age, blood pressure, diabetes, and smoking, to determine whether to start statins. In the study, the risk predicted by the CXR-CVD model closely matched actual heart attacks and strokes in participants.

This was just a preliminary study. More research is needed to validate the effectiveness of the CXR-CVD approach. But the simplicity of this method could offer advantages over ASCVD, which requires a lot of health information.

"The beauty of this approach is you only need an x-ray, which is acquired millions of times a day across the world," said lead author Jakob Weiss, MD, a radiologist affiliated with the Cardiovascular Imaging Research Center at Massachusetts General Hospital and the AI in Medicine program at the Brigham and Women's Hospital in Boston.

What you can do

Such new risk assessment tools aren't yet available. In the meantime, the way to predict your future odds of having a heart attack or stroke is with the American College of Cardiology Disease’s online risk calculator.

This questionnaire determines your 10-year risk of cardiovascular disease based on your:

• Age
• Sex
• Race
• Blood pressure
• Cholesterol levels
• History of diabetes and smoking
• Medication use

Take the results to your primary care doctor or cardiologist, who can recommend lifestyle changes like diet and exercise, and possibly medication, to reduce your cardiovascular risks.

YOU MIGHT ALSO LIKE: How to Prevent a Heart Attack