A collaborative research effort between KAIST and the University of Michigan has led to the development of a digital biomarker that utilizes smartwatch data to predict symptoms of depression.

This innovative technology was validated through a study involving around 800 shift workers, showcasing its effectiveness in predicting mood and depression-related symptoms.

The research team has also created filtering technology that accurately estimates the circadian clock's phase using heart rate and activity data collected from smartwatches.

The COVID-19 pandemic has significantly exacerbated mental health issues worldwide, with approximately one billion people affected by psychiatric conditions, including 1.8 million in South Korea alone.

Traditional methods for monitoring circadian rhythms are often invasive and costly, requiring frequent blood tests or polysomnography, which limits access to mental health treatments.

In response to these challenges, the World Health Organization is exploring new treatment directions for mental illness that take into account the sleep and circadian systems, which play a crucial role in emotional responses and overall mood.

Professor Dae Wook Kim emphasized the importance of applying mathematical models to wearable biometric data for mental health management, envisioning a new approach that could enhance care for socially disadvantaged individuals.

The research aims to establish a continuous, non-invasive mental health monitoring system, potentially transforming mental health care and improving access for those in need.

This new technology leverages wearable devices to collect real-time biometric data, overcoming previous limitations associated with indirect measurements of circadian rhythms.

The study, led by Professors Dae Wook Kim and Daniel B. Forger, highlights the critical role of sleep and circadian systems in mental health, influencing both mood and decision-making.

The research received support from multiple institutions, including KAIST, the US National Science Foundation, and the US National Institutes of Health.

Findings from this study were published in the journal 'npj Digital Medicine' on December 5, 2024, marking a significant advancement in the field of mental health monitoring.