SEOUL, Jan. 16 (Korea Bizwire) — A joint research team from South Korea and the United States has developed a wearable device technology capable of detecting early signs of depression without the need for blood tests.
The breakthrough was announced on January 15 by the Korea Advanced Institute of Science and Technology (KAIST).
Professor Kim Dae-wook’s team from KAIST’s Department of Brain and Cognitive Sciences collaborated with Professor Daniel Forger’s team from the University of Michigan’s Department of Mathematics to create a predictive technology using data collected from smartwatches, such as physical activity and heart rate, to identify depression-related symptoms.
The World Health Organization (WHO) has highlighted the body’s ‘biological clock’ and sleep patterns—which directly influence impulsivity, emotional responses, and decision-making—as promising avenues for mental health treatment.
The brain’s central biological clock, located in the hypothalamus, regulates behavioral and physiological functions by maintaining a 24-hour rhythm. For example, the release of melatonin around 9 p.m. prompts sleep at a consistent time.
Traditionally, accurately measuring the biological clock and sleep state requires costly and invasive procedures, such as extracting blood every 30 minutes overnight to monitor melatonin levels and conducting polysomnography (PSG) to evaluate sleep quality. These hospital-based tests are impractical for outpatient mental health patients and present significant cost barriers.
To address this, wearable devices that monitor heart rate, body temperature, and activity in real-time without spatial constraints have gained attention. However, raw data from these devices alone have limited utility as biological markers.
The research team overcame this challenge by applying time-series analysis to estimate 24-hour fluctuations in melatonin levels. This method allowed them to analyze the circadian rhythm of the biological clock using heart rate and activity data collected from wearable devices.
They further employed digital twin technology to simulate desynchronization between the brain’s central clock and the heart’s peripheral clock.
Such desynchronization, often caused by frequent night shifts or irregular work hours, disrupts hormone systems like dopamine, leading to decreased cognitive ability and happiness.
Collaborating with neuroscience professor Srijan Sen and psychiatry professor Amy Bonar from the University of Michigan, the team conducted a large-scale prospective cohort study involving 800 night-shift workers.
Their findings confirmed that the developed digital biomarker could predict not only next-day mood but also six major symptoms of depression, including sleep disturbances, appetite changes, and concentration issues.
Professor Kim stated, “This non-invasive mental health monitoring technology can be effectively used for managing the mental health of socially vulnerable populations.”
The study was published online in the journal npj Digital Medicine on December 5, 2024.
Kevin Lee (kevinlee@koreabizwire.com)