Related Experiment Video
Updated: Sep 15, 2025
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Blood Biomarker-Based Machine Learning Model for Predicting Cognitive Impairment in Stroke Patients
Yue Zhao1, Daojun Zeng2, Hong Yu2
1Department of Anesthesiology, The Affiliated Hospital, Southwest Medical University, Luzhou, Sichuan Province, China; Anesthesiology and Critical Care Medicine Key Laboratory of Luzhou, Southwest Medical University, Luzhou, Sichuan Province, China; Department of Thoracic Surgery, The Affiliated Hospital, Southwest Medical University, Luzhou, Sichuan Province, China.
Insights
Machine learning models accurately predict cognitive impairment (CI) risk in stroke patients. Logistic regression showed the best performance, identifying key factors for early intervention and improved outcomes.
Area of Science:
- Neurology
- Data Science
- Biostatistics
Background:
- Cognitive impairment (CI) is a frequent complication in stroke survivors, negatively impacting prognosis.
- Early and precise identification of stroke patients at high risk for CI is essential for timely management.
- This study developed machine learning (ML) models to predict CI risk in stroke patients.
Purpose of the Study:
- To develop and evaluate ML models for predicting cognitive impairment risk in stroke patients.
- To identify key clinical and demographic factors associated with CI post-stroke.
- To enable early intervention strategies by accurately assessing CI risk.
Main Methods:
- Utilized data from the China Health and Retirement Longitudinal Study (2011-2018) for 2505 stroke patients.
- Employed Lasso regression and Boruta algorithm for feature selection, identifying ten key variables.
- Developed and compared ten ML algorithms, including logistic regression, XGBoost, SVM, and Random Forest, assessing performance via AUC and decision curve analysis.
Main Results:
- The logistic regression model achieved the highest predictive accuracy with an Area Under the Curve (AUC) of 0.824 (95% CI: 0.794-0.854).
- Decision curve analysis confirmed the logistic regression model's superior clinical utility and net benefit.
- SHapley Additive exPlanations identified education, age, pain, depression, HbA1c, and BUN as significant predictors of CI.
Conclusions:
- ML-based prediction models offer high accuracy for assessing CI risk in stroke patients.
- The developed models facilitate early identification of at-risk individuals, paving the way for timely interventions.
- Accurate CI risk prediction can significantly improve patient outcomes following a stroke.
Background:
Cognitive impairment (CI) is common in stroke patients and is associated with a poor prognosis. Early and accurate identification of high-risk patients is crucial. This study aims to develop a predictive model for CI risk in stroke patients using machine learning (ML).
Methods:
Data were sourced from the China Health and Retirement Longitudinal Study for stroke patients between 2011 and 2018. Lasso regression and the Boruta algorithm were used to select key feature variables. Ten ML algorithms were developed: traditional logistic regression, Extreme Gradient Boosting, Support Vector Machine, k-Nearest Neighbors, Gradient Boosting Machine, Adaptive Boosting, Neural Networks, Light Gradient Boosting Machine, Random Forest, and CatBoost. Model performance was evaluated using the area under the curve and decision curve analysis. The SHapley Additive exPlanations method was employed for model interpretation.
Results:
In total, 2505 stroke patients were included, of whom 779 had CI. Ten key feature variables were selected. Among the prediction models, logistic regression model demonstrated the best performance, with an area under the curve of 0.824 [95% confidence interval: 0.794-0.854)]. Decision curve analysis showed that this model provided the highest net benefit. SHapley Additive exPlanations analysis identified education, age, pain, depression, hemoglobin A1c, and blood urea nitrogen as important predictive factors.
Conclusions:
ML-based prediction models demonstrate high accuracy in assessing the risk of CI in stroke patients, enabling early intervention to improve outcomes.
