Repository logo
 

Integrating Virtual Reality and Artificial Intelligence for Mass Casualty Incident Training

Files

Thesis(10.52 MB)

Date

2025

Authors

Supervisor

Ruan, Ji
David, Parry

Item type

Thesis

Degree name

Doctor of Philosophy

Journal Title

Journal ISSN

Volume Title

Publisher

Auckland University of Technology

Abstract

This thesis explores the integration of Virtual Reality (VR) and Artificial Intelligence (AI) for enhancing pre-hospital triage training in the context of Mass Casualty Incidents (MCIs). Addressing the challenges posed by traditional training methods, which often involve significant costs and logistical complexities due to the infrequent nature of MCIs, this research introduces an innovative VR learning tool designed to realistically simulate emergency scenarios. The primary objectives of this study were to develop the VR learning tool using a Design Science Research methodology and to detail the advanced data analysis methodologies employed in addressing the research questions (RQ1–RQ3.3). By leveraging VR and AI technologies, the tool aimed to provide emergency healthcare professionals with a dynamic and immersive training environment, enabling them to practice and refine their triage skills without the constraints of traditional physical simulations. A significant portion of the research was dedicated to addressing the limitations of current VR interaction methods by prototyping more interactive and lifelike user experiences through advanced VR controllers. This enhancement allowed for deeper and more comprehensive data collection, including audio metrics, thereby facilitating a nuanced understanding of trainees’ performance and engagement within the simulated environment. The development process of the VR learning tool, characterized by the integration of AI techniques and statistical methods, reflected the study’s commitment to both innovation and effective assessment of training outcomes. The experimental phase outlined the preparation, execution, and ethical considerations of implementing the VR training, providing insights into the tool’s potential to quantitatively and qualitatively evaluate emergency response skills. The findings from this study indicated that VR technology can be a supplementary tool in emergency healthcare training, particularly in scenarios involving mass casualty incidents. The analysis of training sessions with 10 participants showed variability in performance during simulated car crash and earthquake scenarios: some participants were quick but less accurate, while others were slower yet more precise, reflecting diversity in emergency response approaches. Survey results showed that participants, predominantly aged 18–24 with varying levels of experience, found the VR training highly immersive and engaging, although some reported physical discomfort, highlighting the need for ergonomic improvements. Additionally, AI-driven analysis of speech data demonstrated improved consistency and accuracy in participants’ communication over time, emphasizing the importance of clear communication in emergencies.

Description

Keywords

Source

DOI

Publisher's version

Rights statement

Permanent link

http://hdl.handle.net/10292/19824

Collections

Doctoral Theses