Predictive modelling of hospital-acquired infection in acute ischemic stroke using machine learning

Abstract

Hospital-acquired infections (HAIs) are serious complication for patients with acute ischemic stroke (AIS), often resulting in poor functional outcomes. However, no existing model can specifically predict HAI in AIS patients. Therefore, we employed the Gradient Boosting matching learning algorithm to establish predictive models for HAI occurrence in AIS patients and poor 30-day functional outcomes (modified Rankin Scale > 2) in AIS patients with HAI by analyzing electronic health records from 6560 AIS patients. Model performance was evaluated through internal cross-validation and external validation using an independent cohort of 3521 AIS patients. The established models demonstrated robust predictive performance for HAI in AIS patients, achieving area under the receiver operating characteristic curves (AUROCs) of 0.857 ± 0.008 during internal validation and 0.825 ± 0.002 during external validation. For AIS patients with HAI, the second model effectively predict poor 30-day functional outcomes, with AUROCs of 0.905 ± 0.009 during internal validation and 0.907 ± 0.002 during external validation. In conclusion, machine learning models effectively identify the HAI occurrence and predict poor 30-day functional outcomes in AIS patients with HAI. Future prospective studies are crucial for validating and refining these models for clinical application, as well as for developing an accessible flowchart or scoring system to enhance clinical practices.

Fig. 1

Distribution and classification of hospital-acquired infections (HAI) in patients with acute ischemic stroke (AIS). (A) Culture positivity rate of HAI in the urinary tract, respiratory tract, bloodstream, and other areas (including soft tissue, joint, intra-abdominal, and nasal sinus). (B) Distribution of Gram-negative pathogens isolated from blood, sputum, and urine in AIS patients with HAI. (C) Distribution of Gram-positive pathogens isolated from blood, sputum, and urine samples from AIS patients with HAI.

Fig. 2

Machine learning model performance and feature importance for predicting hospital-acquired infections (HAIs) in patients with acute ischemic stroke (AIS). (A, B) Receiver Operating Characteristic (ROC) curves of the model showed mean Area under the Curve (AUROC) values of 0.86 (A) for internal validation and 0.82 (B) for external validation. (C) Distribution of Shapley Additive Explanations (SHAP) values for each patient. Red indicated high values of continuous features or “true” for categorical features, while blue indicates low values of “false”. Gray spots represent missing values. A positive SHAP value suggests the feature contributes to predicting HAI. (D) The AUROC of the model trained using the most important features (top 1 to 8) assessed with internal and external validation cohort. All features were assessed on admission. MRC: Medical Research Council.

Fig. 3

Machine learning model performance and feature importance in distinguishing between favorable and poor functional status 30 days after stroke onset, as defined by a modified Rankin scale score greater than 2. (A, B) Receiver Operating Characteristic (ROC) curves of the model showed mean Area Under the Curve (AUROC) values of 0.90 (A) for internal validation and 0.91 (B) for external validation. (C) Distribution of Shapley Additive Explanations (SHAP) values for each patient. Red indicated high values of continuous features or “true” for categorical features, while blue indicates low values of “false”. Gray spots represent missing values. A positive SHAP value suggests the feature contributes to predicting poor functional outcome. (D) The AUROC of the model trained using the most important features (top 1 to 8) assessed with internal and external validation cohort. (MRC: Medical Research Council).