[Interview] NAIRL Korea University Researcher Changhee Lee: “AI Must Read Signals from the Real World”

Changhee Lee, Professor in the Department of Artificial Intelligence at Korea University and a participating researcher in the National AI Research Lab, spoke with JoongAng Ilbo about how AI can read time-varying signals, such as 119 emergency call conversations, neonatal breathing patterns, and medical data, to support decision-making in real-world settings. Professor Lee’s team published a study on real-time AI analysis of 119 emergency calls in npj Digital Medicine this February, showing that AI can more accurately identify the risk of cardiac arrest by considering factors such as a patient’s age, location, breathing condition, and conversational context, even when a caller says the patient is “still breathing.”

The article also highlighted Professor Lee’s work on time-series AI. Time-series AI analyzes data that changes over time, including patients’ heart rate and blood pressure, neonatal breathing, and factory sensor data, to support decision-making in the field. Professor Lee emphasized the importance of developing AI that goes beyond high accuracy by being able to explain why it made a particular judgment and by being usable in real-world settings.

Professor Lee noted that solving real-world problems in areas such as healthcare, emergency response, and manufacturing requires a long-term structure that enables collaboration among researchers, industry, and international partners. With this shared understanding, the National AI Research Lab is building a collaborative foundation that connects researchers and industry partners in Korea and abroad, supporting long-term research on AI that works in real-world environments.

Read more in the article below.
https://www.joongang.co.kr/article/25430770