In a collaborative study between Korea Advanced Institute of Science and Technology (KAIST) and the University of Michigan, researchers have developed a wearable device technology to predict mood changes and depression symptoms.

Mental health disorders affect approximately 1 billion people worldwide. In Korea alone, about 1.8 million people suffer from depression and anxiety disorders, with the total number of mental health patients increasing by 37 percent in the past five years to reach 4.65 million.

The World Health Organization has identified the circadian clock and sleep stages as promising directions for mental health treatment, as these factors directly influence impulsivity, emotional responses, decision-making, and overall mood.

However, traditional methods of measuring these factors require overnight hospital stays and invasive procedures, making them impractical and expensive for most outpatients.

The team, led by Professors Kim Dae-wook of the Department of Brain and Cognitive Sciences at KAIST and Daniel B. Forger of the Department of Mathematics at the University of Michigan, developed a technique to predict symptoms such as sleep disorders, depression, loss of appetite, overeating, and decreased concentration using activity levels and heart rate data collected from smartwatches.

The research team used a filtering technology that accurately estimates the phase of the circadian clock using time-series data from smartwatches. This digital twin of brain circadian rhythms was validated through a large-scale prospective cohort study involving approximately 800 shift workers, in collaboration with Professors Srijan Sen and Amy Bohnert from the University of Michigan.

"This research provides a breakthrough in applying wearable biodata to actual disease management through mathematics," Professor Kim said. "We expect this to present a new paradigm in mental health care, particularly helping vulnerable populations who currently must take active steps to seek help when experiencing depression symptoms."

The findings of the research were published in npj Digital Medicine.