Machine Learning-Based Prognostic Models for Long-Term Care in Traumatic Brain Injury Patients Unable to Return Home: A Retrospective Analysis of ICU Admissions

Abstract

Abstract Background Traumatic brain injury (TBI) research often emphasizes mortality rates or functional recovery, overlooking the critical aspect of long-term care needed by patients reliant on institutional and Respiratory Care Ward (RCW) support. This study employs machine learning techniques to develop predictive models for analyzing the prognosis of this patient group. Method Retrospective data from electronic medical records at Chi Mei Medical Center, encompassing 2020 TBI patients admitted to the ICU between January 2016 and December 2021, were collected. A total of 44 features were included, utilizing four machine learning models and various feature combinations based on clinical significance and Spearman correlation coefficients. Predictive performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve and validated with the DeLong test and SHAP (SHapley Additive explanations) analysis. Result Notably, 236 patients (11.68%) were transferred to long-term care centers. XGBoost with 27 features achieved the highest AUC (0.823), followed by Random Forest with 11 features (0.817), and LightGBM with 44 features (0.813). The DeLong test revealed no significant differences among the best predictive models under various feature combinations. SHAP analysis illustrated a similar distribution of feature importance for the top eleven features in XGBoost with 27 features and Random Forest with 11 features. Conclusion Random Forest demonstrated clinically meaningful predictive capability under 11-feature combinations. This predictive model provides early insights into patients' subsequent care trends, enabling proactive arrangements for institutional or RCW support.