This AI program can predict heart attack, stroke risk within decade using single chest X-ray


CHICAGO— A patient’s risk of a heart attack or stroke in the next 10 years can be predicted from just one chest X-ray by a new computer system.

Researchers with the Radiological Society of North America say it combines AI (artificial intelligence) with a standard X-ray to find patterns associated with hardening of the arteries. The technique offers hope that doctors will be able to prescribe vulnerable individuals cholesterol-reducing drugs before it’s too late.

“Our deep learning model offers a potential solution for population-based opportunistic screening of cardiovascular disease risk using existing chest X-ray images,” says study 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, in a media release.

“This type of screening could be used to identify individuals who would benefit from statin medication but are currently untreated.”

Eating plenty of fish, fruit and vegetables, and going for brisk walks or bike rides are also protective habits that can lower heart disease risk.

Deep learning is a complex series of algorithms that enable machines to make forecasts based on patterns in data. The method, presented at the annual meeting of the RSNA in Chicago, could revolutionize heart therapy.

Who needs heart disease medication?

Current guidelines recommend estimating a patient’s 10-year risk to establish who should take statins for primary prevention. This is based on the ASCVD (atherosclerotic cardiovascular disease) risk score which takes into account a host of factors. They include age, sex, race, high blood pressure, smoking history, Type 2 diabetes, and blood tests. Those scoring 7.5 percent or more should receive statins.

“The variables necessary to calculate ASCVD risk are often not available, which makes approaches for population-based screening desirable,” Dr. Weiss says. “As chest X-rays are commonly available, our approach may help identify individuals at high risk.”

Normal chest X-ray (CREDIT: Radiological Society of North America)

The US team trained the model, known as CXR-CVD risk, to predict death from cardiovascular disease using 147,497 chest X-rays from 40,643 participants in a cancer screening trial.

“We’ve long recognized that X-rays capture information beyond traditional diagnostic findings, but we haven’t used this data because we haven’t had robust, reliable methods,” Dr. Weiss continues. “Advances in AI are making it possible now.”

In tests, the system accurately predicted heart attacks and strokes in a group who had routine chest X-rays at Mass General Brigham. About 10 percent of the 11,430 outpatients suffered a major cardiac event over the average follow-up of just over a decade. The system also identified those who were eligible for potentially life-saving statin therapy.

The new system takes advantage of one of the most common medical scans

“The beauty of this approach is you only need an X-ray, which is acquired millions of times a day across the world,” the study author explains. “Based on a single existing chest X-ray image, our deep learning model predicts future major adverse cardiovascular events with similar performance and incremental value to the established clinical standard.”

If a controlled, randomized trial validates the results it could support doctors in making the right decisions for treatment.

“What we’ve shown is a chest X-ray is more than a chest X-ray,” Dr. Weiss concludes. “With an approach like this, we get a quantitative measure, which allows us to provide both diagnostic and prognostic information that helps the clinician and the patient.”

Cardiovascular disease is the world’s number one killer – claiming almost 18 million lives a year.

South West News Service writer Mark Waghorn contributed to this report.





Source link

Leave a Comment

Your email address will not be published. Required fields are marked *