Kyung Hee University Professor Jinseok Lee's Team Develops AI to Predict Trauma Mortality

Utilizing Korean Data for Emergency Medical Services

Kyung Hee University announced on January 23 that the research team led by Professor Jinseok Lee from the Department of Biomedical Engineering has developed an artificial intelligence (AI) model that can predict the risk of death in trauma patients in real time during the pre-hospital ambulance stage.

The research team developed the AI model using data collected from trauma centers in Korea. Jointly developed with Professor Woosung Kang's team at Gachon University Gil Medical Center, the model predicts the risk of death in the emergency room in real time using only basic information that paramedics can obtain at the scene.

While conventional trauma severity assessment indicators rely on information available only after hospital arrival, this new model quantitatively presents the risk of death even before arrival, enabling medical staff to proactively prepare the necessary personnel and resources for response.

In addition, the model underwent external validation using data from four regional trauma centers in Korea, including Gachon University Gil Medical Center, as well as trauma centers in Australia. As a result, it demonstrated excellent predictive performance not only with domestic data but also with overseas data, where the medical systems and data structures differ.

Notably, the research team has built the model into a web-based real-time system that can be used in actual trauma centers. Furthermore, they plan to expand it into a multimodal emergency medical AI platform that integrates wearable sensors, voice recognition, and image-based AI.

Professor Lee stated, "This is a case that proves the reliability of emergency medical AI technology designed in Korea, even in overseas settings," adding, "Our goal is to expand the scope of validation and apply it in clinical settings to reduce trauma mortality rates."

Meanwhile, this research, supported by the Ministry of Science and ICT, was published this month in the international journal 'Nature Communications.'

This content was produced with the assistance of AI translation services.