One single chest X-ray is all it takes, thanks to Chicago researchers.
At the Radiological Society of North America (RSNA) researchers announced their study, which showed that a deep learning model utilizing a single X-ray could predict a 10-year risk of death from heart attack or stroke. The AI in this study finds patterns within an image that’s associated with further diseases.
To calculate the risk of a 10-year major adverse cardiovascular disease event, an ASCVD risk score is used. Variables include age, blood pressure, sex, race, blood tests, etc. Statin medication is given for primary prevention to those who have a 7.5% or higher 10-year risk.
Jakob Weiss, M.D., stated that many of the variables necessary to calculate ASCVD risk aren’t available, and that chest X-rays “may help identify individuals at high risk”. Dr. Weiss also noted that including a controlled, randomized trial is vital to validating the study’s deep learning model. Using the AI approach, the team has shown they can provide diagnostic and prognostic information that informs both clinician and patient.