Three professors receive funding to examine AI decision support tools to enhance medical triage
Featured experts
- Mandy Agnew, PhD, professor and executive director, Injury Biomechanics Research Center in the School of Health and Rehabilitation Sciences
- Angela Harden, PhD, research assistant professor, Injury Biomechanics Research Center in the School of Health and Rehabilitation Sciences
- Randee Hunter, PhD, associate clinical professor, Injury Biomechanics Research Center in the School of Health and Rehabilitation Sciences
Amanda Agnew, PhD, Angela Harden, PhD and Randee Hunter, PhD received funding from the Ohio Federal Research Network for the study “REACT-X: Resilient Explainable AI for Casualty Triage with Logistics Awareness Using Deep Topological Modeling.” They join research partners from Mined XAI, Wright State University and the Ohio Manufacturing Innovation Center to examine a platform that uses an artificial intelligence (AI) decision support tool which uses deep topological modeling to identify critical features that would enhance medical triage, identify patients requiring urgent life-saving interventions and inform medical logistics needs.
Especially needed in modern military operations that take place in complex, geographically dispersed environments, REACT-X could help medics, physicians and civilians face existing challenges in managing limited resources while responding to and triaging casualty patients. The research’s objectives are to:
- Examine how REACT-X integrates multimodal datasets (physiological signals, injury data, supply utilization records and other metadata) to provide a framework for identifying critical features to aid medics in resource-constrained, contested environments.
- Examine how it extracts information from multimodal datasets that provide XAI-enabled risk assessment, allowing frontline providers to optimize scarce medical supplies, identify patients at considerable risk of decompensation and adjust treatment strategies.
- Examine how this dual-use approach leverages XAI and modular architectures, ensuring transparency, adaptability and compatibility with existing government and commercial platforms.
- Review the effect REACT-X has on helping medics determine when to evacuate patients, how to allocate limited resources and when to request timely resupply across distributed battlespaces.
- Examine how the framework can provide practitioners at various levels of expertise with role-tailored recommendations, enabling more informed, faster, and more effective medical decisions that increase survival rates in prolonged casualty care scenarios.
REACT-X could inform logistics and help medics be prepared to deliver and sustain life-saving care.
