Loading...
Current coronary artery disease (CAD) risk assessment algorithms give categorical estimations of risk (e.g., low, intermediate, high) that do not take into account the continuum of disease severity and risk for death. To define a quantitative marker of CAD that could give a more granular assessment of risk along the spectrum of disease, researchers leveraged a machine-learning model trained on electronic health record (EHR) data.
Using EHR data from ≈96,000 participants of two longitudinal biobank cohorts, they derived, validated, and externally tested the “ISCAD” risk score, which ranges from 0 (lowest probability of CAD) to 1 (highest probability) and utilizes established risk factors, pooled cohort equations, and polygenic risk scores. Fo…