Artificial intelligence predicted the risk of cardiovascular death
Artificial intelligence is able to predict the risk of death from cardiovascular disease and stroke based on a single chest x-ray. The use of such a technique will make it possible to estimate the risk of cardiovascular disease among patients who do not have access to specialized cardiac care.Using deep learning model can predict the 10-year risk of death from heart attack or stroke based on the results of a single chest x-ray. Researchers at the Massachusetts Cardiovascular Imaging Research Center have come to these conclusions. The results of the study, presented at the annual meeting of the Radiological Society of North America RSNA 2022, published Medscape.
Researchers developed an algorithm to identify cardiovascular risk from the results of a single chest radiograph of participants in the large-scale PLCO study. The results of 147,500 radiographs from 40,600 participants were used to train the model.
The ability of artificial intelligence to predict 10-year risk of atherosclerotic cardiovascular disease was compared with the standard method recommended by the American College of Cardiology (ACC) and the American Heart Association (AHA), which takes into account age, sex, race, systolic blood pressure level, treatment for hypertension, smoking, presence of type 2 diabetes and lipidogram.
An independent group of 11,430 patients who underwent outpatient chest radiography was used for comparison. The mean age of the participants was 60 years. During 10.3 years of follow-up, 1,096 patients (9.6%) had serious cardiovascular events.
The likelihood of developing serious cardiovascular events was twice as high among participants in whom the deep-learning program identified the risk of cardiovascular events by radiography. The relationship between the risk identified by artificial intelligence and the development of serious cardiovascular events remained statistically significant even when analyzed with known risk factors (the risk ratio was 1.63).
The probability of atherosclerotic cardiovascular events calculated using the developed model and the standard method was similar.
The scientists noted that the reliability of the developed algorithm should be further confirmed using data from younger patients. The use of such a technique will make it possible to assess the risk of cardiovascular disease among patients who do not have access to specialized cardiac care.