Presentation
Identifying Robust Predictors of Operational Fatigue in Emergency Medical Services Providers: A Machine Learning Approach
DescriptionThis study applied a wearable sensor-based machine learning approach to model operational fatigue among EMS providers under real-world conditions. Using multimodal physiological, behavioral, and contextual data collected across five days, Logistic Regression achieved the best performance (AUROC = 0.87). Robust predictors shared across models, including time of day, cumulative work exposure, dispatched calls, and pulse rate dynamics, indicated the combined influence of circadian rhythm, workload accumulation, and physiological changes in fatigue. Findings support the feasibility of scalable wearable-based fatigue monitoring to enhance fatigue risk visibility in EMS operations.
Event Type
Lecture
TimeTuesday, October 20th3pm - 3:20pm PDT
Location
Similar Presentations