Can Application of Artificial Intelligence Improve Emergency Department Triage Performance?

Area of Science:

• Medical Informatics
• Clinical Decision Support Systems
• Artificial Intelligence in Healthcare

Background:

• Emergency Department (ED) overcrowding is a global issue, leading to delayed care and adverse patient outcomes.
• Current triage systems struggle to identify critically ill patients due to subtle symptoms.
• Artificial intelligence (AI) offers potential to enhance ED clinician capabilities in patient assessment.

Purpose of the Study:

• To systematically review the literature on the effectiveness of AI in improving Emergency Department triage performance.
• To answer the question: "Can the application of artificial intelligence improve Emergency Department triage performance?"

Main Methods:

• A systematic literature review was conducted using PubMed.
• Inclusion criteria focused on studies from the last 5 years, involving human subjects, in English, and using keywords: Artificial Intelligence, Emergency Department, Triage.
• A total of 211 results were initially identified, with 38 articles ultimately reviewed in full.

Main Results:

• The literature search identified 211 relevant articles.
• After initial screening, 62 articles were deemed potentially relevant.
• 38 articles were selected for full review based on their responsiveness to the research question.

Conclusions:

• Emerging evidence suggests AI-based ED triage systems outperform traditional methods.
• These AI systems demonstrate cohort-specific superiority in predicting the need for critical care, surgery, and admission for both adult and pediatric patients.
• AI holds significant potential to optimize patient flow and resource allocation within Emergency Departments.