Cardiovascular diseases, as you know, are the leading cause of death in the world. According to the World Health Organization (WHO), they take around 18 million lives every year: like 10 second world wars in the last 40 years.
Devastating consequences, which have inspired researchers to work to reduce the risk and improve treatment and prevention of heart disease and related diseases.
Among the most advanced solutions, a technology called CXR-CVD. It is an artificial intelligence that using a single chest x-ray, predicts the risk of dying within 10 years from a heart attack or stroke caused by atherosclerotic cardiovascular disease.
AI that “sees” the risk 10 years in advance
The learning model, “trained” in a special study designed by the National Cancer Institute in the United States, is extremely promising. It also takes advantage of a second independent cohort of about 11.430 outpatients, all of whom had chest X-rays that made them potentially eligible for statin therapy, a type of prevention to help with heart attack risk.
The results of the study were presented to the annual meeting of the Radiological Society of North America (RSNA).
“Our deep learning model offers a potential solution for cardiovascular disease risk screening using also existing X-ray images,” says the study's lead author, the radiologist Jacob Weiss, MD
Estimates that will be critical in determining who should take a statin for primary prevention.
How is the risk calculated?
The calculation is done using a statistical model called the “atherosclerotic cardiovascular disease (ASCVD) risk score”. It considers a number of variables including age, gender, race, systolic blood pressure, smoking, type 2 diabetes and blood tests.
If the 10-year risk is equal to or greater than 7,5%, treatment with statins is recommended.
The potential to save millions of lives
Considering that all those variables needed to calculate risk aren't always available, you'll understand that using a simple X-ray can completely transform your life diagnoses.
Advances in artificial intelligence are making this possible now.
