South Korea has begun piloting an AI-powered ambulance emergency support platform designed to improve paramedic documentation, patient assessment and hospital transfer decisions.
Developed by researchers at Yonsei University Severance Hospital in collaboration with the National Fire Agency, the prototype—known as the Intelligence Emergency Activity Support Platform—uses at least 10 AI models to support four core functions.
These include converting paramedic voice communications into structured clinical records, predicting patient deterioration and on-scene risk, generating pre-triage severity scores using ambulance CCTV footage, and providing decision support for treatment guidance and transfer hospital selection.
In field testing, paramedics rated the platform 86 out of 100 in overall satisfaction, citing the hospital transfer recommendation feature as particularly useful for real-world decision-making.
The project has now entered its second phase, which will focus on quantitatively evaluating response speed, workload reduction, communication accuracy and system stability in real emergency settings.
Severance Hospital noted that emergency responders often risk missing the golden hour as they simultaneously assess patients, identify available hospitals and transmit critical information, with post-event documentation increasing the risk of missed details.
The AI platform aims to streamline emergency coordination by enabling faster, more accurate communication between ambulance crews and emergency department clinicians while reducing reporting burden.
The initiative also involves the Electronics and Telecommunications Research Institute and the Korea Electronics Technology Institute.
The pilot aligns with broader healthcare reforms following the lifting of South Korea’s “serious” national medical crisis alert in October, which had been in place for 20 months after the mass resignation of junior doctors in February 2024.
As part of its transition to long-term reforms, the government has capped virtual consultations at 30%, standardised emergency medical fee schedules and advanced multiple AI-led initiatives to modernise emergency care, including patient classification, transfer systems and emergency department decision support tools.
