Development and validation of an explainable machine learning model for risk stratification in patients with clinically suspected acute pulmonary embolism: retrospective study

Abstract

Highlights • A machine learning model predicts acute pulmonary embolism risk in suspected patients.• LASSO and logistic regression identified key clinical predictors from 77 variables.• Eight ML algorithms were compared to select the optimal predictive model.• SHAP framework provides interpretable explanations for individual risk predictions.• The model aims to reduce unnecessary CTPA overuse and radiation exposure.