Innovation through Artificial Intelligence in Triage Systems for Resource Optimization in Future Pandemics.

Area of Science:

• Healthcare technology
• Infectious disease management
• Emergency medicine

Background:

• Artificial intelligence (AI) is increasingly integrated into healthcare.
• AI offers significant potential benefits for hospital emergency services.
• Rapid patient assessment is crucial during pandemics.

Purpose of the Study:

• To develop and evaluate a novel AI-trained system for early detection of patients at risk of infection during pandemics.
• To compare the speed and accuracy of AI-driven triage against standardized systems in emergency departments.
• To facilitate timely implementation of preventive measures to curb disease transmission.

Main Methods:

• A machine learning system, specifically the extreme gradient boosting (XGB) algorithm, was employed for emergency department triage.
• The system analyzed over 89 variables automatically for patients presenting during a pandemic, using the COVID-19 era as a case study.
• Patient data was collected consecutively as they presented to the emergency department.

Main Results:

• The XGB system achieved a high balanced accuracy of 91.61% in identifying patients at risk.
• The AI system provided results significantly faster than traditional triage methods.
• Key predictors of mortality included procalcitonin levels, age, oxygen saturation, LDH, C-reactive protein, chest X-ray findings, and D-dimer levels.

Conclusions:

• The extreme gradient boosting (XGB) algorithm proves to be a valuable and innovative tool for triage systems.
• This AI approach can effectively guide patient care pathways during future pandemics, exemplified by the COVID-19 experience.
• Early risk identification via AI enhances pandemic response and patient management.