Related Experiment Video
Updated: May 2, 2026
Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking
Published on: February 12, 2011
3D Cardiac Magnetic Resonance Substrate Features-based Machine-learning Model for Postmyocardial Infarction Risk
Lujing Wang1, Xiaoying Zhao1, Yuhong Fan2
1Department of Radiology, The Second Affiliated Hospital of Kunming Medical University, Kunming, China.
Background:
Accurate risk stratification postmyocardial infarction (MI) remains challenging. In this study we aimed to develop interpretable machine-learning (ML) models integrating 3-dimensional (3D) cardiac magnetic resonance (CMR) substrate features to predict major adverse cardiovascular events (MACEs) after MI.
Methods:
This retrospective study included MI patients who underwent CMR between May 2015 and October 2024. The primary endpoint was MACE. External validation used multicenter datasets. 3D features (core scar, border zone, abnormal corridors) were extracted via ADAS 3D. Feature selection involved univariate logistic regression and the Boruta algorithm. Eight ML models were trained; the top performer, the tabular prior-data-fitted network (TabPFN), was used to build multimodal models. Shapely Additive Explantations (SHAP) analysis provided interpretability.
Results:
Two hundred ninety-two MI patients were finally enrolled. During 35-month median follow-up, 91 experienced MACEs. Nine key predictors were identified: 3 clinical (high-density lipoprotein, chronic kidney disease, tricuspid regurgitation), 2 functional (left ventricular ejection fraction, left ventricular circumferential strain), and 4 3D substrate features (border zone mass, corridor mass, burden, length). Model 4 (clinical + 3D features) showed strong performance across training (area under the curve [AUC] = 0.91), internal (AUC = 0.82), and external (AUC = 0.89) sets. Model 3 (only 3D features) had an external AUC = 0.90, surpassing clinical (AUC = 0.63) and functional (AUC = 0.49) models. Decision curve analysis highlighted the clinical benefit of incorporating 3D features. SHAP analysis identified corridor mass and burden as key predictors.
Conclusions:
ML models using 3D CMR substrate features significantly improve post-MI MACE prediction compared with traditional methods, offering interpretable and personalized risk stratification tools.
