Machine learning to predict high-risk coronary artery disease on CT in the SCOT-HEART trial.

ABSTRACT:

Background

Machine learning based on clinical characteristics has the potential to predict coronary CT angiography (CCTA) findings and help guide resource utilisation.

Methods

From the SCOT-HEART (Scottish Computed Tomography of the HEART) trial, data from 1769 patients was used to train and to test machine learning models (XGBoost, 10-fold cross validation, grid search hyperparameter selection). Two models were separately generated to predict the presence of coronary artery disease (CAD) and an increased burden of low-attenuation coronary artery plaque (LAP) using symptoms, demographic and clinical characteristics, electrocardiography and exercise tolerance testing (ETT).

Results

Machine learning predicted the presence of CAD on CCTA (area under the curve (AUC) 0.80,

SUBMITTER: Williams MC 

PROVIDER: S-EPMC12406813 | biostudies-literature | 2025 Sep

REPOSITORIES: biostudies-literature