Can a Generative Artificial Intelligence Model Be Used to Create Mass Casualty Incident Simulation Scenarios? A
Sergio M Navarro1, Angie G Atkinson2, Ege Donagay3
1Division of Trauma, Critical Care, and General Surgery, Department of Surgery, Mayo Clinic, 200 1st St. SW, Rochester, MN 55905, USA.
Healthcare (Basel, Switzerland)
|December 30, 2025
Summary
Generative artificial intelligence (AI) can create mass casualty incident (MCI) simulation scenarios. Combining AI generation with expert human review enhances scenario quality for training, though human oversight is crucial.
Area of Science:
- Medical Simulation and Training
- Artificial Intelligence in Healthcare
- Trauma Care Education
Background:
- Mass casualty incident (MCI) simulation scenarios are critical for multidisciplinary trauma team preparedness.
- Current scenario development is resource-intensive, relying on detailed planning by expert teams.
- Assessing the feasibility of generative AI for creating these complex training tools is essential.
Purpose of the Study:
- To evaluate the feasibility of using generative artificial intelligence (AI) for developing mass casualty trauma simulation scenarios.
- To assess the quality and reliability of AI-generated scenarios compared to traditional methods.
- To explore a novel workflow integrating AI with human expertise for simulation development.
Main Methods:
- Ten complex MCI trauma simulation scenarios were generated using a large language model (LLM) platform (ChatGPT4) based on public US trauma data.
- Scenarios were evaluated by two Advanced Trauma Life Support (ATLS) certified raters using the validated Simulation Scenario Evaluation Tool (SSET).
- LLM scenarios were compared to two control scenarios; iterative human revision and re-evaluation were performed.
Main Results:
- Initial LLM-generated scenarios scored a median of 78.5 on SSET, significantly lower than control scenarios (median 94).
- High interrater reliability (ICC 0.965) was observed for initial LLM scenarios.
- Post-revision, LLM scenarios achieved a median SSET score of 94 with improved interrater reliability (ICC 0.7425).
Conclusions:
- A collaborative workflow combining LLM generation with expert human review shows promise for creating MCI simulation scenarios.
- LLMs offer a scalable solution for developing trauma training materials, enhancing efficiency.
- Essential human oversight, quality assurance, and governance are vital for ensuring clinical accuracy and educational value.
Related Concept Videos
01:18Steps in Outbreak Investigation
466
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
466
01:16Non-equilibrium in the Cell
5.3K
An important concept in studying metabolism and energy is that of chemical equilibrium. Most chemical reactions are reversible. They can proceed in both directions, releasing energy into their environment in one direction, and absorbing it from the environment in the other direction. The same is true for the chemical reactions involved in cell metabolism, such as the breaking down and building up of proteins into and from individual amino acids, respectively. Reactants within a closed system...
5.3K