On December 18, Professor Yoo Seung-chan of Yonsei University's Department of Biomedical Systems and Information Sciences will explain the main functions of Severance Hospital's next-generation electronic medical record system, “Y-NOT.'' Photo courtesy of Asia Today
December 28th (Asia Today) — A research team at Yonsei University has built a generative AI-based medical records system, currently being used at Severance Hospital, with the goal of reducing the amount of time doctors spend creating documentation and allowing them to focus more on patient care.
The system, known as Y-NOT and implemented through the hospital's Y-NOT recording platform, uses a large-scale language model to create draft admission and discharge records that clinicians can review, said Yoo Seung-chan, a professor at Yonsei University's School of Biomedical Systems and Information Sciences.
“Why do we spend more time sitting in front of a computer than examining patients?” Yu said, explaining a question frequently asked by emergency medical staff who helped drive the project.
Yu said that full-scale development began last year as the administrative burden on medical workers increased amid rising tensions between doctors and the government. He said the team decided that AI technology was mature enough to significantly reduce record-keeping workloads and began development.
Yu said the project began in July last year and was in clinical practice by November, with model development and hospital deployment occurring in parallel. He said the team focused first on the clinical utility that could be achieved, rather than pursuing only maximum performance of the model.
Some medical staff initially expressed concerns, including the risk of errors in records, questions about responsibility for errors, and concerns that the system would infringe on doctors' authority. Mr Yu said two surveys conducted after implementation showed that these concerns were alleviated, with high levels of satisfaction, especially among older staff.
He said the team built the system to support clinicians rather than replace them. Doctors will continue to diagnose and make decisions, he said, while AI will draft and organize documents for verification. Yu added that some staff say the system has made care easier by reducing the need to manually search past electronic records.
Yoo said the Y-NOT system is now being used for broader record-keeping beyond the emergency department, including operating rooms and inpatient wards.
The time needed to create emergency room medical records was cut by more than half, from 69.5 seconds to 32 seconds, the report said. Staff say the reduced paperwork burden gives them more time to make eye contact with patients, Yu said.
He said the evaluation showed improved record completeness and standardization across the care team, including nurses, and the time savings were helping emergency physicians move on to the next patient more quickly or provide additional guidance to patients being discharged.
Yoo said the long-term goal extends beyond documentation tools to an intelligent agent system designed to support safe care in line with global standards. He said that direction is tied to Severance Hospital's Doctor Answer 3.0 project, and future plans include exploring ways for patients to communicate with AI systems based on their medical records.
Yoo said AI could help address the growing demand for healthcare as the population ages and the number of essential medical staff decreases. He said this will support clinicians' guideline-based care and allow patients to maintain an ongoing connection with the hospital after they are discharged.
— Asia Today reported. Translation by UPI
© Asia Today. Unauthorized reproduction and redistribution is prohibited.
