AI for Smarter Emergency Care: Predicting the Right Facility for Every Patient

Overview

Emergency care decisions can be complex, especially when patients are unsure whether to visit an Emergency Department (ED) or a Minor Illness and Injury Clinic (MIIC). This collaborative research project is developing AI-driven tele-triage tools to guide patients to the most appropriate facility based on their self-reported symptoms.

Building on earlier work that identified challenges in patient self-assessment, the team led by PhD researcher Hetiao (Slim) Xie, is now leveraging large language models (LLMs) to interpret symptom descriptions and predict the best care pathway. The study uses anonymised clinical data from both ED and MIIC settings to train and validate the model.

This research demonstrates how enterprise-scale AI can be responsibly applied in healthcare. By intelligently routing patients based on acuity, the model improves safety for high-risk cases while reducing ED overcrowding through efficient MIIC referrals. More broadly, it showcases the integration of clinical expertise, human-centred design, and data science to build intelligent systems that enhance decision-making, optimise resource allocation, and strengthen public trust in digital health technologies.