Development and Validation of a Web-Based Machine Learning Model for Predicting Early Neurological Deterioration Following Stroke Thrombolysis: Multicenter Study

Abstract

Background: Early neurological deterioration (END) significantly worsens outcomes in patients with acute ischemic stroke (AIS) receiving intravenous thrombolysis, yet clinicians lack reliable tools to identify high-risk patients who need intensified monitoring and preemptive interventions.

Objective: This study aimed to develop and validate a high-performance machine learning model for END prediction that enables personalized risk-stratified management of patients with AIS after thrombolysis.

Methods: This multicenter study analyzed 1927 patients with AIS who were treated with intravenous thrombolysis in 3 hospitals, comprising a development cohort (n=1361) from Lianyungang Clinical Medical College and an external validation cohort (n=566) from 2 independent hospitals. We systematically evaluated 27 clinical parameters using multiple machine learning algorithms to develop ENDRAS (Early Neurological Deterioration Risk Assessment Score), a prediction model based on 6 readily available clinical variables. Model performance was assessed through comprehensive metrics (area under the receiver operating characteristic curve, accuracy, precision, recall, F1-score) in both internal and external validation cohorts.

Results: The XGBoost-based ENDRAS showed promising predictive performance (area under the receiver operating characteristic curve=0.988, 95% CI 0.983-0.993) using 6 readily available parameters: Trial of ORG 10172 in Acute Stroke Treatment classification, intracranial artery stenosis severity, National Institutes of Health Stroke Scale score, systolic blood pressure, neutrophil count, and red blood cell distribution width. We established a dual-pathway management protocol for stratifying patients into low-risk (<29%) and high-risk (≥29%) groups, where high-risk patients receive intensive monitoring with hourly assessments and expedited imaging, while low-risk patients follow a resource-optimized protocol without compromising safety. Implemented as a web-based calculator with a <0.02-second computation time, ENDRAS enables real-time clinical decision support at the point of care.

Conclusions: ENDRAS integrates END prediction into actionable clinical pathways, potentially improving postthrombolysis care through personalized monitoring strategies and targeted interventions. Its robust performance in merged cohorts, efficient computation time, and structured management framework address key challenges in stroke care while enhancing resource utilization. Further prospective validation across diverse populations is needed to fully establish ENDRAS as a standard clinical decision-support system, but its ability to identify high-risk patients early may significantly improve outcomes in AIS.

Keywords: acute ischemic stroke; clinical decision support; early neurological deterioration; intravenous thrombolysis; machine learning; prediction model; stroke.

Figure 1.. Patient enrollment flow diagram. AUC: area under the receiver operating characteristic curve; END: early neurological deterioration; ENDRAS: Early Neurological Deterioration Risk Assessment System; IVT: intravenous thrombolysis; RFECV: recursive feature elimination with cross-validation; SHAP: Shapley additive explanations; SMOTE: synthetic minority oversampling technique.

Figure 2.. Computational efficiency of machine learning algorithms. CPU: central processing unit; DT: decision tree; LR: logistic regression; MLP: multilayer perceptron; NB: naive Bayes; RF: random forest; SVM: support vector machine.

Figure 3.. Area under the receiver operating characteristic curves (AUC) of the XGBoost model on the training set, internal validation set, and external validation set.

Figure 4.. Comparison of performance metrics (area under the receiver operating characteristic curves [AUC], accuracy, F₁-score) of the XGBoost model across 4 datasets (training set, internal validation set, external validation set, and merged dataset), presented as a radar chart.

Figure 5.. Comparison of performance metrics (area under the receiver operating characteristic curves [AUC], accuracy, F₁-score) of the XGBoost model across 4 datasets (training set, internal validation set, external validation set, and merged dataset), presented as a bar chart.

Figure 6.. Feature importance for risk prediction by the XGBoost model. Ccr: creatinine clearance rate; DBP: diastolic blood pressure; DNT: door-to-needle time; FB: fibrinogen; HbA_1c: glycosylated hemoglobin; HCY: homocysteine; IAS: intracranial atherosclerotic stenosis; INR: international normalized ratio; LAA: large-artery atherosclerosis; NEUT: neutrophil; NIHSS: National Institutes of Health Stroke Scale; ODT: onset-to-door time; RDW: red cell distribution width; SBP: systolic blood pressure; TOAST: Trial of ORG 10172 in Acute Stroke Treatment; WBC: white blood cell.

Figure 7.. Evaluation of the confusion matrix of the combined dataset using the Early Neurological Deterioration Risk Assessment System (ENDRAS) model.

Figure 8.. Receiver operating characteristic curves of Trial of ORG 10172 in Acute Stroke Treatment (TOAST) large-artery atherosclerosis (LAA), intracranial atherosclerotic stenosis (IAS), National Institutes of Health Stroke Scale (NIHSS), systolic blood pressure (SBP), neutrophil (NEUT), red cell distribution width (RDW), and Early Neurological Deterioration Risk Assessment System (ENDRAS).

Figure 9.. Calibration curve of the Early Neurological Deterioration Risk Assessment System (ENDRAS) model.

Figure 10.. Decision clinical curve of the Early Neurological Deterioration Risk Assessment System (ENDRAS) model.