Related Experiment Video
Updated: Jul 6, 2026
A Test Bed to Examine Helmet Fit and Retention and Biomechanical Measures of Head and Neck Injury in Simulated Impact
Published on: September 21, 2017
Medical documentation using artificial intelligence (AI) for battlefield injury simulations: A comparative study
Daniel Gelman1, Danielle Akler, Gil Shimon
1Trauma and Operational Medicine Division, Surgeon General Headquarters, Israel Defense Forces Medical Corps, Tel Hashomer, Ramat Gan, Israel (D.G., D.A., G.S., S.O., I.K., M.L., O.A., A.B., Z.A.B., R.N.); Department of Military Medicine, Faculty of Medicine, Hebrew University, Jerusalem, Israel (D.G., I.K., O.A., Z.A.B.); and The Azrieli Faculty of Medicine, Bar-Ilan University, Safed, Israel (A.B.).
Background:
Incomplete prehospital documentation remains a major challenge, compromising continuity of care and data quality. Manual documentation is time-consuming and cognitively demanding, particularly under operational stress. Speech-based artificial intelligence (AI) systems may enable real-time documentation without interrupting clinical workflow.
Methods:
We conducted a feasibility study comparing AI-assisted speech-based documentation with conventional manual documentation during simulated prehospital trauma care. Eight teams completed trauma scenarios using both documentation methods in randomized order. AI outputs were structured into 40 predefined documentation fields derived from the standard combat casualty card. Documentation completeness and user-reported usability were compared between methods. Analyses were performed on paired team-scenario observations.
Results:
Forty-four simulations were completed (22 per method), with successful AI output in 91% of cases. Overall documentation completeness was similar between AI-assisted and manual documentation [73.9% (interquartile range, 64.1-83.7) vs. 64.8% (interquartile range, 58.7-79.2); p =0.31], with no differences across scenarios or documentation categories (all p >0.05). Usability ratings favored the AI-assisted approach for overall ease of use ( p =0.024) and data entry ( p =0.015).
Conclusions:
In this simulated prehospital battlefield setting, AI-assisted speech-based documentation was feasible and achieved documentation completeness comparable to manual documentation while improving perceived usability. Further development and real-world evaluation are needed for operational implementation. ( J Trauma Acute Care Surg . 2026;101: S89-S94. Copyright © 2026 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Association for the Surgery of Trauma.).
Level Of Evidence:
Diagnostic Test/Criteria; Level III.