A domestic research team has released an algorithm technology developed for sleep health management to users around the world through the recently launched Galaxy Watch 8, including Samsung Galaxy Watches.
Professor Kim Jae-kyung of the Korea Advanced Institute of Science and Technology (KAIST) Department of Mathematical Sciences, who is also a group leader at the Institute for Basic Science (IBS) Life Science Mathematics Group, noted on the 28th that their self-developed personalized sleep guide algorithm will be made available to global users through a collaboration with Samsung Electronics.
This technology is a sleep time recommendation algorithm developed based on mathematical modeling and circadian rhythm theory, which analyzes users' sleep patterns to suggest optimal bedtime ranges. The Galaxy Watch uses sensors such as an accelerometer and heart rate sensor to detect users' movements, heart rate, and sleep stages, and precisely analyzes the sleep patterns. The algorithm suggests the optimal bedtime based on this data, tailored to individuals' circadian rhythms. Instead of simply recommending sleep volume, it provides quantitative and actionable information such as 'it is ideal to go to bed between 11:10 p.m. and 11:40 p.m.' based on individuals.
While existing smartwatch sleep functions primarily focused on analyzing past data, such as 'how many hours did you sleep last night,' the research team's sleep guide algorithm designs the future of sleep. It is like not just telling the weather from yesterday but forecasting tomorrow's weather and advising to carry an umbrella.
Professor Kim Jae-kyung said, 'While we have been independently developing a sleep health app and steadily researching and improving it for nearly three years, there were many trial and errors as a non-professional development team, and commercialization was not easy.' He added, 'With the help of Professor Seo Moon-jong from the KAIST Technology Value Creation Center, we were able to collaborate with Samsung Electronics and implement the developed technology broadly into real-world services.'
The algorithm was selected for the Hot Topics session at the world's largest sleep conference, SLEEP 2025, held in the United States last June, and is scheduled to be introduced at the World Sleep 2025 conference in Singapore in September.
Professor Kim is currently collaborating with Professor Joo Eun-yeon from Samsung Medical Center to develop a more advanced sleep time recommendation algorithm. Recently, they have also publicly released a sleep disorder prediction algorithm called 'SLEEPS' through joint research. Additionally, they are concurrently working on the commercialization of the sleep app within their research lab.