Clinical Report: Predicting Early Neurological Deterioration in Acute Ischemic Stroke
Overview
This study developed a multidimensional machine learning model to predict early neurological deterioration (END) in acute ischemic stroke (AIS) patients. The model identified key predictors and demonstrated superior performance in risk stratification, potentially aiding timely clinical interventions.
Background
Acute ischemic stroke (AIS) is a leading cause of disability and mortality, with early neurological deterioration (END) occurring in 10-40% of patients. END is associated with poor long-term outcomes and highlights the need for effective prediction tools. Traditional methods often fail to capture the multifactorial nature of END, necessitating advanced predictive models.
Data Highlights
Model
AUC (Training Set)
AUC (Validation Set)
Random Forest
0.779
0.775
K-Nearest Neighbors
0.727
0.741
Gradient Boosting Machine
0.736
0.665
Key Findings
Five core predictors for END were identified: NIHSS score, blood glucose, infarct core volume, collateral circulation status, and NLR.
Admission NIHSS score, blood glucose, infarct core volume, and NLR were independent risk factors for poor prognosis.
Collateral circulation status was identified as an independent protective factor.
The Random Forest model outperformed KNN and GBM in predictive performance.
The model provides a practical tool for early identification of high-risk END patients.
Clinical Implications
The findings suggest that incorporating multidimensional clinical data can enhance the prediction of END in AIS patients. Clinicians may utilize this model to identify high-risk patients early, facilitating timely interventions that could improve outcomes.
Conclusion
The multidimensional machine learning model represents a significant advancement in predicting END in AIS patients, potentially guiding clinical decision-making and improving patient management.
