AI model predicts CVD risk from single chest radiograph
A recently developed AI model has been shown to predict ten-year incidences with a similar level of success to the current standard of care.
The AI model uses data from a single chest X-ray to predict the ten-year risk of death from incidences such as heart attacks or strokes, following atherosclerotic cardiovascular disease (ASCVD). The results of this study were recently presented at the Radiological Society of North America (RSNA) conference 2022.
Risk assessment is a crucial part of the prevention of ASCVD, so a new method of detecting this risk ten years in advance appears to be a hugely positive step in identifying high-risk patients, giving them the possibility for various preventative and curative therapies. Currently, risk assessments rely on certain risk factors: age, sex, systolic blood pressure, antihypertensive treatment, total and HDL-cholesterol, diabetes and smoking.
The study developed the AI model using 147,497 chest radiographs from 40,643 participants from the PLCO cancer screening trial, the AI model was trained to predict CVD mortality from a single X-ray image.
Jakob Weiss, a radiologist associated with the Cardiovascular Imaging Research Center at Massachusetts General Hospital who was also involved in this study, commented that “variables necessary to calculate ASCVD risk are often not available, which makes approaches for population-based screening desirable.” He continued that “as chest X-rays are commonly available, our approach may help identify individuals at high risk.”