KAIST and the University of Michigan Research Team “Predicting Depression in 800 Shift Workers”
(Daejeon = Yonhap News) Reporter Park Joo-young = A joint research team from Korea and the United States has developed wearable device technology capable of detecting early signs of depression without blood tests in hospitals.
The Korea Advanced Institute of Science and Technology (KAIST) announced on the 15th that Professor Dae-Wook Kim’s team from the Department of Brain and Cognitive Sciences, in collaboration with Professor Daniel Forger’s team from the Department of Mathematics at the University of Michigan, has developed a technology to predict symptoms related to depression using data like activity levels and heart rate collected from smartwatches (wristwatch-type wearable devices).
The World Health Organization (WHO) is paying attention to ‘biological clocks’ and ‘sleep’, which have direct impacts on impulsivity, emotional reactions, and decision-making, as promising directions for treating mental disorders.
The biological clock (central clock) located in the pituitary gland of the brain regulates behavioral and physiological phenomena in our body by maintaining a consistent 24-hour rhythm.
The secretion of the melatonin hormone in the brain at 9 PM, prompting a person to sleep at a regular time, is also due to the biological clock.
For accurate measurement of the biological clock and sleep state, melatonin hormone concentration changes in the body must be measured every 30 minutes overnight, and a polysomnography test (PSG – evaluating sleep quality through brain waves, respiration, heart rate, etc., during sleep) must be performed.
This limits testing to hospitalization, making it challenging for outpatients with mental disorders and with high testing costs.
Therefore, wearable devices that measure biological data like heart rate, body temperature, and activity levels in real-time without spatial constraints are gaining attention, but the data alone is challenging to use as actual biomarkers (indicators of changes in the body).
The research team succeeded in analyzing the circadian rhythm of the biological clock by utilizing a time-series analysis technique to estimate the 24-hour change in melatonin hormone concentration based on heart rate and activity data measured by wearable devices.
Through Digital Twin technology (implementing the machinery, equipment, objects, etc., of the real world in the virtual world), the team simulated the desynchronization between the central clock in the brain and the peripheral clock of the heart.
Desynchronization between the central clock and sleep-wake cycles caused by frequent night shifts or shift work leads to disturbances in the hormone system, including dopamine, resulting in issues such as reduced cognitive ability and a decrease in well-being.
The research team collaborated with Professor Srijan Sen from the Neuroscience Research Institute and Professor Amy Bonner from the Department of Psychiatry at the University of Michigan for a large-scale prospective cohort study involving 800 nighttime shift workers.
The study confirmed that the digital biomarker developed by the research team could predict not only mood for the next day but also six symptoms indicative of depression, including sleep problems, appetite changes, and decreased concentration.
Professor Dae-Wook Kim stated, “The non-invasive mental health monitoring technology developed in this study could be utilized for mental health management in socially vulnerable groups.”
This research outcome was published in the online version of the international journal ‘npj Digital Medicine’ on December 5th of the previous year.