Driving Change in Hospitals Through Medical AI

- MOHW holds the 2025 Medical AI Training Program Outcomes Sharing Conference -

- Sharing best practices in AI adoption and discussing directions for advancing medical AI training in 2026 -

The Ministry of Health and Welfare (MOHW, Minister Jeong Eun Kyeong) held the Outcomes Sharing Conference for the 2025 Medical AI Training Program for Healthcare Professionals on Friday, December 5, at the Grand Mercure Imperial Palace Seoul Gangnam.

This program was launched in June 2025 under a supplementary budget to align with the Lee Jae Myung administration’s “National AI Strategy.” Its purpose is to bridge the AI adoption gap in clinical settings and build a foundation for the safe expansion of medical AI. Organized by the Korea Human Resource Development Institute for Health & Welfare (KOHI), the program is being implemented by four medical institutions: Samsung Medical Center, Seoul National University Hospital, Asan Medical Center, and Chung-Ang University Gwangmyeong Hospital.

Approximately 100 participants attended the conference, including MOHW Vice Minister Lee Hyung-hoon, KOHI Acting President Bae Nam-young, representatives from the implementing institutions, distinguished trainees, and subject-matter experts. Participants shared best practices and practical clinical improvements achieved at each institution, while discussing the program’s direction for 2026 and identifying areas for policy enhancement.

Outstanding institutions and trainees received awards from the Minister of Health and Welfare and the KOHI President. Key outcomes and best practices are as follows:

   * Two Minister’s Awards and four KOHI President’s Awards were presented.

Asan Medical Center operated hands-on AI projects utilizing hospital data and developed AI agents* based on Large Language Models (LLMs) and medical foundation models. Through these efforts, the center explored solutions to clinical challenges, such as optimizing clinical protocols and building smart chunking–based RAG** systems. They also showcased projects leveraging diverse datasets, including a medical dictionary–based RAG system.

   * AI agent: An AI system capable of understanding a goal, executing necessary tasks, and solving problems autonomously.

  ** Retrieval-Augmented Generation (RAG): A method in which an AI system supplements its internal knowledge by retrieving information from external documents or databases to generate more accurate outputs.

At Chung-Ang University Gwangmyeong Hospital, 16 teams participated in an AI Prompt-thon*, where multidisciplinary teams jointly identified issues, designed LLM prompts**, and executed the full lifecycle from prompt development to application. This initiative produced several examples of workflow innovation using data from the Health Insurance Review and Assessment Service (HIRA), including a RAG model for predicting reductions in insurance reimbursements.

   * Prompt-thon: A compound of “prompt” and “marathon,” referring to a competition where participants develop new ideas and create products or services within a set timeframe using generative AI.

  ** Prompt: A command or inquiry given to an AI system to direct its actions.

Samsung Medical Center established a model for regional expansion by partnering with local hospitals, such as Kangbuk Samsung Hospital and Changwon Samsung Hospital. Through medical AI practicums and advanced short-term training, the center strengthened the AI competencies of regional medical institutions. Seoul National University Hospital assessed organizational AI readiness and acceptance through pre- and post-training surveys and on-site interviews, proposing policy recommendations and internal strategies to support the adoption of medical AI.

During the panel discussion, implementing institutions, instructors, and trainees discussed key field issues under the theme “Changes Driven by Medical AI Training and Policy Priorities for Improvement.” Topics included bridging AI infrastructure gaps across hospitals, the necessity of standardized training, and strategies for inter-hospital collaboration.

Vice Minister Lee Hyung-hoon stated, “Changes in clinical practice begin with strengthening the capabilities of frontline healthcare professionals. This year, we laid the groundwork for medical AI workforce training and confirmed the need for continued investment. In 2026, we will actively support the digital transformation of hospitals and the wider adoption of medical AI, ensuring that tailored training is provided according to each institution’s size and environment.”

He further emphasized, “We will develop this program into a leading medical AI initiative in which healthcare professionals take the lead in addressing on-site challenges—spanning training, consulting, development, implementation, and advancement.”

- END -