Changi General Hospital and IBM Build AI-Powered Emergency Training Platform for Paediatric Crisis Response

Singapore's public healthcare system has been quietly building some of the most practical AI applications in Southeast Asia — not in flashy consumer tools, but in the high-stakes environment of emergency medicine. The latest example comes from Changi General Hospital (CGH), which is partnering with IBM to develop an AI-powered training platform purpose-built for paediatric mass-casualty incidents, an area where traditional drills have long fallen short.

Announced at IBM Think Singapore 2026, the platform combines mixed-reality simulations with IBM's watsonx.ai and watsonx Orchestrate software to run real-time scenario exercises. At its core is an AI Assessor component that analyses voice communications, treatment decisions, and coordination patterns during simulated emergencies, then generates structured feedback for post-exercise reviews. The system tracks how clinicians communicate, delegate, and make treatment decisions under pressure — human factors that paper-based exercises and manikin drills simply cannot measure at scale.

The collaboration addresses a known gap in emergency preparedness: existing drills are resource-intensive, difficult to scale, and limited in the performance data they produce. A single mass-casualty simulation can involve dozens of staff across multiple departments, yet the debrief relies heavily on subjective observation. CGH's platform aims to make assessments consistent and repeatable across teams, so the hospital can run complex emergency-preparedness programmes more frequently and across a broader range of scenarios.

Adjunct Assistant Professor Jimmy Goh, a senior consultant in CGH's Department of Emergency Medicine, is careful to frame the technology as augmentative rather than autonomous. "The human decision, it always comes first. The AI helps to support the role of the human physician, helps to support the decision-making, makes information available faster, more accurately," he said during the media briefing. Goh also cautioned against overreliance on the technology, noting that the goal is to optimise efficiency in decision-making without allowing thinking processes to become so dependent on AI "that it becomes a crutch."

Why it matters for Singapore: This project represents a concrete example of Singapore's healthcare AI strategy moving beyond administrative tools into clinical training — a space where the margin for error is near zero. CGH, which runs more than 70 AI projects through its dedicated AI Office, is effectively serving as a testbed for how public hospitals can integrate AI into mission-critical workflows. If the platform proves effective, the model could extend to other emergency departments across Singapore's public healthcare cluster, raising the baseline for crisis response training nationwide.