Prediction of the 180 day functional outcomes in aneurysmal subarachnoid hemorrhage using an optimized XGBoost model

Conventional models are unable to fully assess the complexity of aneurysmal subarachnoid hemorrhage (aSAH). In this study, we developed a predictive model using the extreme gradient boosting (XGBoost) algorithm to guide individualized treatment by combining inflammatory markers and clinical grading. The study was retrospectively analyzed using 264 patients with aSAH admitted to Yongchuan Hospital of Chongqing Medical University from January 2020 to December 2022 as a training cohort and 88 patients admitted from January 2023 to December 2023 as an external validation cohort. Patients were categorized into favorable and unfavorable prognosis groups based on 6-month modified Rankin Scale (mRS) scores. Significant predictors identified by multivariate logistic regression included NAR, procalcitonin, CRP, D-dimer, and modified Fisher scores. The model had an AUC of 0.87 and a Brier score of 0.13; the validation cohort had an AUC of 0.85 and a Brier score of 0.18. Decision curve analyses underscored the consistent net benefit of the model across thresholds. This study emphasizes the value of integrating clinical and laboratory markers in prognostic prediction and supports targeted interventions to reduce risk and improve prognosis in patients with aSAH.