Related Experiment Video
Updated: Jun 21, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Establishment of the China Elderly Comorbidity Medical Database (CECMed) and its application in machine
Jingwen Shi1, Duanchang Wan2, Wen Tang1
1Department of Geriatrics, Beijing Friendship Hospital, Capital Medical University, Beijing, China.
Insights
Machine learning models effectively predict in-hospital adverse events in elderly Chinese patients with comorbidities. The eXtreme Gradient Boosting (XGBoost) model showed superior performance in identifying high-risk individuals.
Area of Science:
- Geriatric Medicine
- Computational Health
- Data Science in Healthcare
Background:
- Comorbidity is a significant health challenge and leading cause of mortality in China's elderly population.
- Developing predictive tools for adverse events is crucial for improving geriatric care and outcomes.
Purpose of the Study:
- To establish a multicenter dataset focused on geriatric comorbidities in China.
- To evaluate the efficacy of various machine learning models for early warning of in-hospital adverse events.
Main Methods:
- Collected data from elderly individuals across northern, central, and southern China.
- Developed a specialized geriatric comorbidity dataset.
- Applied and compared machine learning models including Random Forest, SVM, 1D CNN, GBDT, and XGBoost for adverse event prediction.
Main Results:
- Over 90% of patients exhibited at least one geriatric syndrome.
- Gradient Boosting Decision Tree (GBDT) and eXtreme Gradient Boosting (XGBoost) demonstrated strong predictive performance (AUROC > 0.90).
- The Shapley Additive Explanation (SHAP) method identified key predictive features.
Conclusions:
- A valuable geriatric comorbidity dataset has been successfully established.
- XGBoost emerged as the top-performing model for predicting in-hospital adverse events.
- Frailty, D-dimer, disease severity, Barthel Index, and fibrin degradation products are significant risk factors.
Aims:
Comorbidity is highly prevalent in the elderly in China, representing a leading cause of mortality in this population. This study established a multicenter dataset specific to geriatric comorbidities and explored the performance in early warning of in-hospital adverse events using multiple machine learning models.
Methods And Results:
Data were collected in the elderly from northern, central, and southern regions of China. Following data processing, a dataset specific to geriatric comorbidities was established. Among the patients, over 90% had at least one geriatric syndrome. Machine learning methods were applied to predict adverse events during hospitalization, including Random Forest, Support Vector Machine (SVM), 1-Dimensional Convolutional Neural Network (1D CNN), Gradient Boosting Decision Tree (GBDT), and eXtreme Gradient Boosting (XGBoost). GBDT (AUROC = 0.91, ACC = 0.903, REC = 0.878, PRE = 0.808, F1 = 0.836) and XGBoost (AUROC = 0.914, ACC = 0.91, REC = 0.893, PRE = 0.817, F1 = 0.848) demonstrated better prediction performance. Shapley Additive Explanation (SHAP) method was used to identify features significantly associated with the occurrence of adverse events and presented the top ten features based on their significance.
Conclusions:
A geriatric comorbidity-specific dataset was established. XGBoost demonstrated the better performance in predicting risk of in-hospital adverse events. Frailty, D-dimer, disease severity grade, Barthel Index, and fibrin degradation products were significantly associated with the occurrence of such events.