Related Experiment Video
Updated: Mar 12, 2026

Estimate the Cognitive Load Using Electrocardiographic Measure: A Human-AI Collaborative Task
Published on: December 5, 2025
Artificial intelligence for battlefield triage in large-scale combat operations: Opportunities, limits, and ethical
Quentin Mathais1, Pierre-Julien Cungi, Antoine Lamblin
1Federation of Anesthesiology and Intensive Care, Sainte Anne Military Teaching Hospital, Toulon (Q.M., P.J.C., J.B.); Department of Anesthesiology and Intensive Care, Laveran Military Teaching Hospital, Marseille (A.L.); French Armed Forces Health Academy, Paris (A.L., O.D., J.B.); Aix-Marseille Université, CNRS, EFS, ADES (A.L.); Anesthesiology and Intensive Care Department, Timone Hospital (S.B.); Laboratoire de Biomécanique Appliquée, LBA UMR T24, Université Gustave Eiffel-Aix-Marseille Université, Marseille, France (S.B.).
Abstract:
Large-scale combat operations (LSCOs) impose major constraints on battlefield medical systems, combining sustained casualty inflow, degraded communications, prolonged evacuation timelines, and limited opportunities for repeated clinical reassessment. Under such conditions, conventional triage frameworks-designed for episodic assessment and rapid evacuation-become insufficient. This narrative review examines how artificial intelligence (AI) could support battlefield triage in LSCO, not as a replacement for clinical judgement, but to preserve situational awareness and prioritization over time when human vigilance alone is insufficient. Based on military medical, technological, and doctrinal literature, we analyze AI through three operational functions: extending caregiver perception, sustaining cognition under pressure, and enabling anticipatory and personalized decision-making. Near-term deployable capabilities include wearable physiological sensors, early warning systems, digital casualty documentation, and unmanned platforms supporting remote assessment, resupply, and evacuation coordination. Mid-term developments may integrate multimodal data fusion, predictive decision support, augmented reality-assisted guidance, and partial automation of prioritization. Longer-term conceptual frameworks, such as digital twins, envision fully predictive and individualized triage and resource allocation but remain at the research stage. We further examine the engineering, human, doctrinal, ethical and strategic constraints that govern AI deployment in LSCO, including DDIL environments, data quality, cognitive and ergonomic risks, automation bias, survivability concerns in a transparent battlefield and requirements for robust governance. Overall, the value of AI for triage in LSCO lies in human-machine teaming that sustains vigilance, coordination, and anticipation under extreme operational constraints, provided deployment remains disciplined, ethically governed, and operationally grounded. ( J Trauma Acute Care Surg . 2026;101: S177-S187. Copyright © 2026 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Association for the Surgery of Trauma.).