Related Experiment Video
Updated: Mar 3, 2026
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
Development and validation of a machine learning predictive model for one-month post-revascularization angina in
Jincheng Wang1, Conghui Zhou2, Bihua Tang2
1Institute of Literature in Chinese Medicine, Nanjing University of Chinese Medicine, Nanjing, China.
Insights
A new machine learning model accurately predicts post-revascularization angina (PRA) risk using key patient factors. This tool aids in personalized care after procedures like percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG).
Area of Science:
- Cardiology
- Machine Learning
- Predictive Analytics
Background:
- Recurrent angina after coronary revascularization (PCI/CABG) presents clinical challenges and increases healthcare costs.
- Existing risk tools have limited accuracy in predicting short-term recurrence.
- Post-revascularization angina (PRA) significantly impacts patient quality of life.
Purpose of the Study:
- To develop and validate a machine learning (ML) model for predicting PRA.
- To identify key clinical factors associated with PRA risk.
- To improve early risk stratification for patients undergoing coronary revascularization.
Main Methods:
- Utilized data from 626 patients in a derivation cohort and 127 in an external validation cohort across Chinese clinical centers (2016-2018).
- Employed the Boruta algorithm for feature selection and trained eight ML models, including Random Forest (RF).
- Validated models internally and externally using metrics like AUC, accuracy, sensitivity, and specificity; SHAP values assessed interpretability.
Main Results:
- The Boruta algorithm identified six key predictors: NYHA class, cardiac troponin T (cTnT), prothrombin time (PT), depression severity, abdominal circumference, and diastolic blood pressure (DBP).
- The Random Forest (RF) model demonstrated superior performance with an AUC of 0.90 (internal validation) and 0.87 (external validation).
- SHAP analysis confirmed that higher NYHA class, elevated cTnT, and depression severity were significant positive predictors of PRA risk.
Conclusions:
- The developed RF model provides a robust and interpretable tool for early PRA risk stratification.
- The model integrates cardiac, hemostatic, psychological, and metabolic factors for comprehensive risk assessment.
- Further prospective, multi-ethnic validation is recommended to enhance the generalizability of the predictive model.
Background:
Recurrent angina pectoris following coronary revascularization via percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) poses significant clinical challenges, associated with reduced quality of life and increased healthcare burden. Traditional risk tools have limitations in predicting short-term recurrence. This study aimed to develop and validate a machine learning (ML) predictive model for post-revascularization angina (PRA).
Methods:
This study used patient data from 38 clinical research centers in 23 provinces of China from 2016 to 2018. Data from 626 patients in a derivation cohort recruited from 28 centers across 16 Chinese provinces and 127 in an external validation cohort from another 10 centers across 10 provinces were analyzed. The Boruta algorithm selected key features, and eight ML models were trained on 70% of the derivation cohort, internally validated on 30%, and externally validated. Performance metrics included area under the curve (AUC), decision curve analysis (DCA), accuracy, sensitivity, specificity, and F1 score. The Shapley Additive explanation (SHAP) values provided model interpretability.
Results:
The Boruta algorithm selected six features: New York Heart Association (NYHA) classification, cardiac troponin T (cTnT), prothrombin time (PT), depression severity, abdominal circumference, and diastolic blood pressure (DBP). The Random forest (RF) model outperformed others, achieving an AUC of 0.90 (accuracy 0.88, sensitivity 0.77, specificity 0.92, F1 0.78) in internal validation and 0.87 in external validation. The SHAP algorithm confirmed the features' predictive importance, with higher NYHA class, elevated cTnT, and depression severity positively influencing PRA risk.
Conclusions:
This RF model offers a robust, interpretable tool for early PRA risk stratification, integrating cardiac, hemostatic, psychological, and metabolic factors. It supports personalized post-revascularization care, though prospective, multi-ethnic validation is needed to enhance generalizability.
Related Concept Videos
01:27Coronary Artery Disease V: Interprofessional Care
01:26Angina IV: Management
01:20Angina V: Nursing Management