A new study has shown that biosignal analysis technology in Samsung Electronics' Galaxy Watch, combined with artificial intelligence (AI), can predict "vasovagal syncope" with approximately 85% accuracy.
Chung-Ang University Gwangmyeong Hospital announced Monday that a research team led by Professor Cho Jun-hwan of the Division of Cardiology, in collaboration with Samsung Electronics (005930.KS), has become the first in the world to successfully predict the likelihood of fainting using a smartwatch.
Vasovagal syncope is a condition in which a temporary malfunction of the autonomic nervous system causes a sudden drop in blood pressure and heart rate, leading to loss of consciousness. While it rarely poses a direct threat to life, unexpected episodes can result in secondary accidents such as falls, fractures, and head injuries. The condition can be particularly dangerous for elderly individuals or those driving, operating machinery, or working at heights, which has driven ongoing demand for technology that can detect warning signs early.
The research team conducted a Head-Up Tilt Test on 132 patients with suspected vasovagal syncope while they wore the Galaxy Watch6. The Head-Up Tilt Test is a standard procedure for evaluating vasovagal syncope, in which a patient is tilted at a specific angle while changes in blood pressure and heart rate are monitored. The team used AI machine learning algorithms to analyze continuous heart rate variability (HRV) data collected through the photoplethysmography (PPG) sensor built into the Galaxy Watch6.
The analysis successfully predicted signs of vasovagal syncope with 84.6% accuracy approximately five minutes before onset. Advance prediction of fainting episodes allows patients to secure response time for prevention, such as assuming a safe posture or drinking fluids. It also increases the chances of preventing accidents such as fractures or brain hemorrhages from sudden falls by enabling patients to call for help. The findings demonstrate the potential for the Galaxy Watch to evolve into a personalized health management platform that analyzes users' heart rate patterns and autonomic nervous system changes in real time and issues alerts warning of imminent syncope risk. The study is drawing particular attention because it achieved meaningful predictive performance using only a commercially available smartwatch in everyday settings, without requiring expensive hospital equipment or invasive sensors.
"This study shows that a smartwatch, which anyone can use, can go beyond being a simple wearable device to become an early warning system that protects personal health," Cho said. "Through additional validation in various environments and real-life conditions, we plan to develop this into a technology that can provide practical help to patients in their daily lives."
