First-24-hour machine learning for 30-day mortality prediction in ICU trauma patients: development in MIMIC-III and cross-database evaluation in MIMIC-IV
Preprint
Abstract
ICU trauma patients are clinically heterogeneous, and early mortality risk stratification may support monitoring and resource allocation. We developed machine learning models for 30-day mortality prediction using information recorded during the first 24 hours after ICU admission. In MIMIC-III, six feature configurations were trained using 3,411 patients and compared in a patient-level configuration-selection hold-out subset of 853 patients. The selected XGBoost configuration yielded an area under the precision–recall curve (AUPRC) of 0.556 and an area under the receiver operating characteristic curve (AUROC) of 0.863. For cross-database evaluation, a 228-predictor harmonized XGBoost model was refitted on the complete MIMIC-III cohort and evaluated in 13,747 MIMIC-IV ICU stays without using MIMIC-IV outcomes for model development or recalibration. It achieved an AUPRC of 0.495, an AUROC of 0.825, and a Brier score of 0.109. Calibration was monotonic but showed increasing overprediction at higher predicted risks. First-24-hour clinical information retained predictive value across MIMIC database versions, although internal configuration selection, model differences, same-center provenance, and incomplete feature-mapping documentation limit generalizability and deployment readiness.Keywords
This publication has 20 references indexed in Scilit:
- Retrospective machine learning approach for forecasting in-hospital death in ICU patients after cardiac arrestInformatics in Medicine Unlocked, 2026
- Global, regional, and national burden of injuries, and burden attributable to injuries risk factors, 1990 to 2019: results from the Global Burden of Disease study 2019Public Health, 2024
- Author Correction: MIMIC-IV, a freely accessible electronic health record datasetScientific Data, 2023
- Shock index as a predictor for mortality in trauma patients: a systematic review and meta-analysisEuropean Journal of Trauma and Emergency Surgery, 2022
- From local explanations to global understanding with explainable AI for treesNature Machine Intelligence, 2020
- XGBoost2016
- MIMIC-III, a freely accessible critical care databaseScientific Data, 2016
- Glasgow Coma Scale Motor Score and Pupillary Reaction To Predict Six-Month Mortality in Patients with Traumatic Brain Injury: Comparison of Field and Admission AssessmentJournal of Neurotrauma, 2015
- Assessing the Performance of Prediction ModelsEpidemiology, 2010
- The relationship between Precision-Recall and ROC curves2006