What question did this study set out to answer?

To develop and validate an interpretable model predicting in-hospital mortality from initial noncontrast head CT scans in patients with gunshot wounds to the head.

March 12, 2026

AI-Aided Triage for GSWH: Validating an Interpretable HCT-Based Mortality Model.

Key Points

To develop and validate an interpretable model predicting in-hospital mortality from initial noncontrast head CT scans in patients with gunshot wounds to the head.
Retrospective cohort at a level I trauma center
Included patients with gunshot wounds to the head who underwent HCT
Stratified random split for development and test sets
Multiple-instance learning algorithm utilized for risk prediction
Performance quantified through partitioning and attention visualizations.
47.8% of included patients survived to discharge; 52.2% died
High discrimination for mortality with AUC of 0.92
Sensitivity of 0.88 and specificity of 0.87 at optimal operating point
Attention maps highlighted critical brain structures related to high mortality risk

Abstract

Civilian gunshot wounds to the head (GSWH) carry high mortality yet lack standardized, imaging-based triage tools. Because initial noncontrast head computerized tomography (HCT) is universally obtained but not leveraged with validated, rapid, and reproducible methods, we developed and evaluated an interpretable, attention-based multiple-instance learning (MIL) model to predict in-hospital mortality from the initial HCT. In a retrospective cohort at a single level I trauma center (May 1, 2018-October 31, 2023), we included consecutive adults (≥16 years) with GSWH who underwent HCT, excluding those dead on arrival or without HCT. Of 222 patients, 106 (47.8%) survived to discharge and 116 (52.2%) died. We used a stratified random split to create a development set (n = 168, 75.7%) and an independent test set (n = 54, 24.3%); the development set was repeatedly partitioned 100 times into training and validation subsets to quantify performance uncertainty, and each of the 100 models was evaluated once on the test set. The MIL algorithm produced a prognostic severity score with case-level interpretability via attention maps. On the independent test set, discrimination for mortality was high (area under the curve: 0.92, 95% CI: 0.87-0.94) with sensitivity 0.88 (95% CI: 0.78-0.97) and specificity 0.87 (95% CI: 0.74-0.96) at the optimal operating point. Attention visualizations consistently highlighted brainstem, deep midline, and ventricular injury in high-mortality predictions, aligning with established high-risk neuroanatomy. These findings demonstrate that an interpretable, HCT-based MIL model can deliver objective, reproducible risk estimates and transparent case-level explanations, supporting early prognostication and imaging-first triage in penetrating brain injury.

Ask AI

Mark Helpful

Bookmark

Relay