Early prediction of incident delirium in traumatic brain injury: a multicenter validated and interpretable machine learning approach.

Overview

1. Graduate School of PLA General Hospital, PLA General Hospital, Beijing, China
2. Department of Medical Innovation and Research, PLA General Hospital, Beijing, China
3. Graduate School of Capital Medical University, Capital Medical University, Beijing, China
4. Department of Neurosurgery, Beijing Fangshan District Liangxiang Hospital, Beijing, China
5. Department of Medical Psychology, Ninth Medical Center of PLA General Hospital, Beijing, China

Abstract

Objective: This study aims to develop and externally evaluate a machine learning (ML)-based predictive model for incident delirium in patients with traumatic brain injury (TBI).

Methods: Patients diagnosed with TBI from the MIMIC-IV and eICU-CRD databases were included. Predictors were selected using Boruta and LASSO regression. Five ML algorithms were developed and compared, with logistic recalibration applied to the external cohort. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). Shapley Additive Explanations (SHAP) was utilized to decode individual risk contributions. Subgroup and sensitivity analyses were conducted to define clinical boundaries and evaluate model robustness.

Results: A total of 915 TBI patients from the MIMIC-IV database and 317 from the eICU-CRD database were included. Random Forest (RF) model achieved balanced performance with an internal AUC of 0.819 and an external AUC of 0.706. The model exhibited favorable internal calibration, adequate external recalibration, and positive clinical net benefits (internal: 0.155, external: 0.080). Overall SHAP analysis identified invasive ventilation, Glasgow Coma Scale (GCS), extracranial injury, Acute Physiology Score III (APSIII), hemoglobin and mixed intra-/extra-axial injury as primary predictors. Crucially, stratified SHAP analysis identified invasive ventilation as the primary driver across all strata, with baseline GCS scores attaining their maximum predictive weight in the medium-risk tier. Subgroup analyses of the external cohort indicated robust generalization in younger patients (AUC = 0.780) and those with extracranial injuries (AUC = 0.762), with expected attenuation in subgroups with higher clinical severity (AUC: 0.578–0.589). Sensitivity analyses confirmed the model's stable performance against competing mortality and missing data (all DeLong test p > 0.05).

Conclusions: The RF model demonstrated acceptable discriminative capacity and clinical utility for early delirium prediction in patients with TBI. Supported by SHAP, it translated complex predictions into an actionable three-tiered framework, serving as a valuable adjunct for guiding early monitoring and neuroprotective strategies.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper links to its data, not to its authors' code: see the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none, so it has no map.

Data

Datasets cited

• doi:10.13026/c2wm1r, at the source; found in “Data availability statement”
• doi:10.13026/kpb9-mt58, at the source; found in “Data availability statement”
• physionet.org/content/eicu-crd, at PhysioNet; found in “Data availability statement”
• physionet.org/content/mimiciv, at PhysioNet; found in “Data availability statement”

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: 1. MIMIC-IV (Medical Information Mart for Intensive Care-IV) Repository Name: PhysioNet Direct Link: https://physionet.org/content/mimiciv/ Accession Number (DOI): 10.13026/kpb9-mt58 (Version 3.1) Project Website: https://mimic.mit.edu 2. eICU-CRD (eICU Collaborative Research Database) Repository Name: PhysioNet Direct Link: https://physionet.org/content/eicu-crd/ Accession Number (DOI): 10.13026/C2WM1R (Version 2.0) Project Website: https://eicu-crd.mit.edu.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

• Funding: added National Natural Science Foundation of China: 72404275; National Key Research and Development Program of China: 2020YFC2003402

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 8 authors, 5 keywords, 44 references.

Cite

This paper

Li, C., Zhang, T., Chen, H., Wang, S., Wang, L., Huan, Y., Liu, J., & Liu, L. (2026). Early prediction of incident delirium in traumatic brain injury: a multicenter validated and interpretable machine learning approach. Frontiers in neurology, 17, 1848730. https://doi.org/10.3389/fneur.2026.1848730

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1016/j.isci.2026.116135 [code]

Predicting ICU in-hospital mortality from text-encoded structured EHR data using adaptive transformer layer fusion.

Journal: iScience

In common: physionet.org/content/eicu-crd, physionet.org/content/mimiciv, clinical / translational, 1 reference

[2] doi:10.3389/fncom.2026.1824898

Deep learning guided propofol ketamine dosing and inflammation trajectories in elderly burns.

Journal: Frontiers in computational neuroscience

In common: physionet.org/content/eicu-crd, physionet.org/content/mimiciv, 1 reference

[3] doi:10.1186/s12879-026-13704-w

Association of Staphylococcus aureus bloodstream infection with 7-day incident delirium in critically Ill adults: a propensity-weighted competing risk analysis.

Journal: BMC infectious diseases

In common: DOI 10.13026/kpb9-mt58, physionet.org/content/mimiciv, clinical / translational, other condition

[4] doi:10.1038/s41598-026-57082-4

Glucose-to-potassium ratio has a non-linear J-shaped association with acute kidney injury risk in traumatic brain injury.

Journal: Scientific reports

In common: physionet.org/content/mimiciv, traumatic brain injury, clinical / translational

[5] doi:10.3389/fneur.2026.1849021

Early pharmacologic venous thromboembolism prophylaxis and in-hospital mortality in ICU patients with intracerebral hemorrhage: a landmark, propensity-weighted cohort study.

Journal: Frontiers in neurology

In common: physionet.org/content/mimiciv, clinical / translational

[6] doi:10.1038/s43856-026-01817-x [code]

Visual prompt engineering for multimodal and irregularly sampled medical data.

Journal: Communications medicine

In common: physionet.org/content/mimiciv, clinical / translational

[7] doi:10.1038/s41467-026-71791-4

Dysregulation of macrophage lipid metabolism underlies intracellular bacterial neuroinvasion.

Journal: Nature communications

In common: physionet.org/content/mimiciv

[8] doi:10.3390/bioengineering13070794

Mamba-Bi-LSTM with SHAP-Guided Iterative Refinement for Multimodal ARDS Diagnosis: A Dual-System Framework.

Journal: Bioengineering (Basel, Switzerland)

In common: 2 references

[9] doi:10.3390/s26165109

Feature Selection Analysis and Robustness to Missing Data for Sensor-Based Prognostic Modeling of Alzheimer's Disease Progression.

Journal: Sensors (Basel, Switzerland)

In common: clinical / translational, 1 reference

[10] doi:10.1038/s41598-026-64405-y [code]

Evaluating clinical and neuroimaging predictors for cognitive-behavioral therapy outcome in obsessive-compulsive disorder.

Journal: Scientific reports

In common: clinical / translational, other condition, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.