Related Experiment Video
Updated: Sep 19, 2025
08:22
Isolation and Flow Cytometric Assessment of Neuroimmune Interactions in a Mini-Stroke Murine Model
Published on: June 20, 2025
150
Predicting Stroke-Associated Pneumonia in Acute Ischemic Stroke: A Machine Learning Model Development and Validation
Mengqi Xie1, Zhiying Liu1, Fangfang Dai1
1The Second Clinical Medical College of Xinjiang Medical University, Xinjiang Uygur Autonomous Region, People's Republic of China.
International Journal of General Medicine
|June 18, 2025
Summary
Machine learning accurately predicts stroke-associated pneumonia (SAP) risk. Key factors include immune markers and patient demographics, enabling early clinical decision-making with a new prediction tool.
Area of Science:
- Neurology
- Infectious Diseases
- Computational Biology
Background:
- Stroke-associated pneumonia (SAP) is a severe complication following ischemic stroke.
- SAP significantly increases patient morbidity and mortality.
- Early identification of patients at high risk for SAP is crucial for improved outcomes.
Purpose of the Study:
- To identify key risk factors for stroke-associated pneumonia (SAP).
- To develop and validate a machine learning (ML) model for early SAP risk stratification.
- To create a practical tool for clinical decision support.
Main Methods:
- Retrospective analysis of 574 ischemic stroke patients.
- Training and testing of nine machine learning models using 10-fold cross-validation.
- Evaluation of model performance using accuracy, AUC-ROC, and F1-score; predictor importance assessed via SHAP analysis.
Main Results:
- SAP incidence was 32.4% in the study cohort.
- The LightGBM algorithm showed superior predictive performance without overfitting.
- Top predictors identified included Monocyte-to-lymphocyte ratio (MLR), Systemic Immune-Inflammation Index (SII), NIHSS score, age, Aggregate Index of Systemic Inflammation (AISI), and Platelet-to-lymphocyte ratio (PLR).
Conclusions:
- Machine learning models, particularly LightGBM, demonstrate high accuracy in predicting SAP risk.
- The identified risk factors provide valuable insights into SAP pathogenesis.
- An interactive web-based prediction tool offers clinicians actionable insights for real-time decision-making.
