Related Experiment Video
Updated: Sep 14, 2025
Predicting Amputation using Local Circulating Mononuclear Progenitor Cells in Angioplasty-treated Patients with Critical Limb Ischemia
Published on: September 22, 2020
Optimized feature selection and advanced machine learning for stroke risk prediction in revascularized coronary
Yong Si1, Armin Abdollahi1, Negin Ashrafi1
1University of Southern California, Los Angeles, CA, USA.
Insights
Machine learning accurately predicts stroke risk after coronary revascularization. The CatBoost model, using fewer features, improves upon existing methods for better clinical decisions.
Area of Science:
- Cardiovascular Medicine
- Medical Informatics
- Machine Learning in Healthcare
Background:
- Coronary artery disease (CAD) is a leading cause of mortality globally.
- Stroke is a significant complication following coronary revascularization (PCI/CABG).
- Existing machine learning (ML) models predict postoperative outcomes but lack quantitative stroke risk prediction post-revascularization.
Purpose of the Study:
- To develop and validate ML models for predicting stroke risk in CAD patients undergoing revascularization.
- To address the gap in ML-based quantitative stroke risk assessment.
- To enhance clinical decision-making and improve patient outcomes.
Main Methods:
- Extracted data for 5,757 patients from the MIMIC-IV database.
- Employed feature selection (Pearson, LASSO, ridge, elastic net) reducing 35 to 14 features.
- Evaluated multiple ML models (Logistic Regression, XGBoost, Random Forest, AdaBoost, Naive Bayes, KNN, CatBoost) using AUC-ROC and accuracy on training, testing, and validation sets.
Main Results:
- The CatBoost model achieved superior performance with an AUC of 0.8486 (test set) and 0.8511 (validation set).
- SHAP analysis identified Charlson Comorbidity Index (CCI), length of stay (LOS), and treatment type as key predictors.
- The model showed a 9% improvement in predictive performance over existing literature with a more parsimonious feature set.
Conclusions:
- Integrated feature selection streamlined the predictive model for efficiency and reliability.
- The CatBoost model is proposed for accurate postoperative stroke prediction in CAD patients undergoing revascularization.
- The model provides valuable insights for informed clinical decisions and stroke risk mitigation.
Background:
Coronary artery disease (CAD) remains a leading cause of global mortality, with stroke constituting a significant complication among patients undergoing coronary revascularization procedures, such as percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG). Previous research has demonstrated the successful application of machine learning (ML) in predicting various postoperative outcomes, including poor prognosis following cardiac surgery and the risk of postoperative stroke. Despite these advancements, a critical gap persists in studies quantitatively linking the risk of postoperative stroke to revascularization using ML-based approaches. This study aims to address this gap by developing and validating ML models to predict the risk of stroke in CAD patients undergoing coronary revascularization, with the ultimate goal of enhancing clinical decision-making and improving patient outcomes.
Methods:
We developed an ML framework to predict stroke risk in patients with CAD undergoing revascularization. A total of 5,757 patients were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Feature selection was performed using a combination of Pearson correlation analysis, least absolute shrinkage and selection operator (LASSO), ridge regression, and elastic net. Initially, 35 features were identified based on expert opinion and a comprehensive literature review; the integrated results of the feature selection methods reduced the feature set to 14. The dataset was randomly divided into training, testing, and validation subsets with proportions of 70%, 15%, and 15%, respectively. Several ML models were evaluated, including logistic regression, XGBoost, random forest, AdaBoost, Bernoulli naive Bayes, k-nearest neighbors (KNN), and CatBoost. Model performance was assessed using the area under the receiver operating characteristic curve (AUC-ROC), accuracy, and 500 bootstrapped 95% confidence intervals (CIs) to ensure robust evaluation.
Results:
The CatBoost model demonstrated superior performance, achieving an AUC of 0.8486 (95% CI: 0.8124-0.8797) on the test set and 0.8511 (95% CI: 0.8203-0.8793) on the validation set. Shapley Additive Explanations (SHAP) analysis identified the Charlson Comorbidity Index (CCI), length of stay (LOS), and treatment types as the most influential predictors. Notably, compared to the best existing literature, which reported an AUC of 0.760 on the test set, our model exhibited a 9% improvement in predictive performance while utilizing a more parsimonious feature set.
Conclusion:
By integrating four feature selection methods, we significantly streamlined the feature set, resulting in a more efficient and reliable predictive model. We propose the CatBoost model for the prediction of postoperative stroke in patients with CAD undergoing coronary revascularization. With its high accuracy, the proposed model offers valuable insights for medical practitioners, enabling informed decision-making and the implementation of preventive measures to mitigate stroke risk.
