A single chest X-ray can be used to predict the 10-year risk of dying from a heart attack or stroke due to atherosclerotic cardiovascular disease using a deep learning model that scientists have built. The model offers a potential solution for population-based opportunistic screening of cardiovascular disease risk using existing chest X-ray images.
The study’s lead author, Jakob Weiss, M.D., a radiologist affiliated with the Cardiovascular Imaging Research Center at Massachusetts General Hospital, said, “This type of screening could be used to identify individuals who would benefit from statin medication but are currently untreated.”
The atherosclerotic cardiovascular disease (ASCVD) risk score, a statistical model that considers a wide range of factors, such as age, sex, race, systolic blood pressure, hypertension therapy, smoking, Type 2 diabetes, and blood tests, is used to determine this risk. Patients with a 10-year risk of 7.5% or greater are advised to take statin medication.
Dr. Weiss said, “The variables necessary to calculate ASCVD risk are often not available, which makes approaches for population-based screening desirable. As chest X-rays are commonly available, our approach may help identify high-risk individuals.”
Scientists trained the model using a single chest X-ray (CXR) input. They developed a CXR-CVD risk model to predict the risk of death from cardiovascular disease using 147,497 chest X-rays from 40,643 participants in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial, a multi-center, randomized controlled trial designed and sponsored by the National Cancer Institute.
Dr. Weiss said, “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. Advances in AI are making it possible now.”
The model was tested on a second independent sample of 11,430 outpatients who underwent a regular outpatient chest X-ray at Mass General Brigham and were possibly eligible for statin medication (mean age, 60.1 years; 42.9% male).
Over the course of the median follow-up of 10.3 years, 1,096 patients, or 9.6% out of 11,430 patients, experienced a serious adverse cardiac event. Significant correlations existed between recorded major cardiac events and the risk predicted by the CXR-CVD risk deep learning model.
Additionally, researchers compared the model’s predictive value to the accepted clinical criterion for determining statin eligibility. Due to missing data (such as blood pressure and cholesterol) in the computerized record, this could only be estimated for 2,401 patients (21%). The CXR-CVD risk model performed comparably to the accepted clinical standard for this subset of patients and even added value.
Dr. Weiss said, “The beauty of this approach is you only need an X-ray, which is acquired millions of times a day across the world. 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.”
“Additional research, including a controlled, randomized trial, is necessary to validate the deep learning model, which could ultimately serve as a decision-support tool for treating physicians.”
“What we’ve shown is a chest X-ray is more than a chest X-ray. 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.”
Results of the study were presented today at the annual meeting of the Radiological Society of North America (RSNA).