Augmented reality (AR) is an emerging technology with immense potential for enhancing human-to-human interaction tasks, particularly in high-risk environments such as mass casualty incident (MCI) tri-age. However, developing practical and effective AR tools for this purpose necessitates a meticulous user-centered design (UCD) process, thoughtfully crafted and validated through iterative testing with first responders in increasingly contextually relevant simulations. In academic circles, the perceived complexity and time requirements of such a process might discourage its adoption within the constraints of traditional publishing cycles. This is likely due, in part, to a lack of representative applied UCD examples. This work addresses this challenge by presenting a scholarly UCD framework tailored specifically for MCI triage, which progresses seamlessly from controlled, context-free laboratory settings to virtual patient simulations and finally to realistic patient (actor) scenarios. Moreover, MCIs and triage are under-served areas, likely due to their high intensity and risk. This means developers need to 'get it right' as quickly as possible. UCD and evaluation alone are not an efficient means to developing these complex and dangerous work domains. Thus, this research also delves into a cognitive work analysis, offering a comprehensive breakdown of the MCI triage domain and how those findings inform future AR sup-ports. This analysis serves to fortify the foundation for future UCD endeavors in this critical space. Finally, it is imperative to recognize that MCI triage fundamentally involves human-to-human interaction supported by AR technology. Therefore, UCD efforts must encompass a diverse array of study stimuli and participants to ensure that the technology functions as intended across all demographic groups. It is established that racial bias exists in emergency room triage, creating worse outcomes for patients of color. Consequently, this study also investigates the potential impact of racial biases on MCI triage efficacy. This entire body of work has implications for UCD evaluation methodology, the development of future AR support tools, and the potential to catch racially biased negative performance before responders ever hit the field. / Doctor of Philosophy / Augmented reality (AR) is uniquely situated to make work within high-risk work environments, like mass casualty incident (MCI) response, safer and more effective. This is because AR augments the user's reali-ty with context-relevant information, like by providing a temperature gauge for firefighters that is always in their visual field. Development of such AR tools for a sensitive arena like MCIs requires several rigor-ous steps before those tools can be deployed in the field. It is crucial to engage in a user-centered design (UCD) process in partnership with actual emergency responders so they can help us understand what help they need. We outline that UCD process in Chapter 2. Once we understand what responders say they need help with, we then need to evaluate those pinch points in the broader context of their work. This means that we evaluate how their job process creates the situation where responders need the kind of help they are asking for. Understanding this helps us create solutions that address the responder's needs while we minimize any new problems created with implementing a new tool into the job. What we learned from examining the work domain is described in Chapter 3. Once we have this firm foundational understanding of responder needs and work and we have designed an AR support tool, we need to evaluate that tool for effectiveness. It is too dangerous to put the AR tool straight into the field, so Chapter 4 explores how we can create simulations of an MCI scenario to study our AR support tool. Finally, after evaluating our AR tool within the scenarios and the scenarios themselves, we evaluate (in Chapter 5) other facets of the job that may be impacting MCI response. In our case, we explore how racial bias may be impacting patient care. It is important to study bias as it has implications for future MCI training and AR tool development. Perhaps future work can explore an AR tool that offsets bias-based performance, or a training that helps catch bias before responders ever get to the real field.
| Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/121096 |
| Date | 09 September 2024 |
| Creators | Nelson, Cassidy Rae |
| Contributors | Industrial and Systems Engineering, Gabbard, Joe L., Ivory, James Dee, Oberdorfer, Sebastian, Lau, Nathan Ka Ching |
| Publisher | Virginia Tech |
| Source Sets | Virginia Tech Theses and Dissertation |
| Language | English |
| Detected Language | English |
| Type | Dissertation |
| Format | ETD, application/pdf |
| Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.0283 seconds