Single-person households in Korea surpassed 8 million, accounting for 36% of all households, a record high. A Seoul city survey found that 62% of single-person households said they feel "lonely," indicating worsening isolation and mental health problems.
Amid this, researchers at Korea Advanced Institute of Science and Technology (KAIST) developed a technology that can precisely track mental health status by using Internet of Things (IoT) data in the home. They said mental health deterioration can be detected early through "changes in daily rhythms" that are hard to capture with smartphones or wearables.
The study is expected to lay the foundation for developing a personalized mental health management system.
KAIST on Jan. 21 announced results from a four-week field study led by Professor Lee Eui-jin of the School of Computing that targeted 20 single-person households among young adults. The team collected IoT data from home appliances, sleep mats, and motion sensors installed in the home and analyzed it together with smartphone and wearable data.
Existing mental health tracking methods are based on smartphone usage or wearables, but data are missing at home if the devices are not carried. In contrast, IoT data in the home reflected users' life patterns seamlessly, enabling more accurate analysis.
The study found a clear tendency for levels of depression, anxiety, and stress to rise as sleep time decreased. It also identified a correlation in which higher indoor temperatures were associated with worse anxiety and depression.
Participants' behavior patterns varied. They were categorized into types such as the "binge-eating type," with increased refrigerator use under stress, and the "lethargic type," with a sharp drop in activity. But there was a commonality that mental health worsened as daily rhythms became more irregular.
The researchers noted that "variability in daily patterns is a more important factor than the frequency of specific behaviors," emphasizing that a regular routine is key to maintaining mental health.
Participants directly checked their life data through a visualization program, and many felt the data helped them understand their mental health more than they worried about privacy invasion. As a result, satisfaction with participation in the study rose significantly.
Lee said, "This study showed that IoT data in the home can be an important clue to understanding mental health in the context of an individual's daily life," adding, "We plan to use artificial intelligence (AI) to predict individual life patterns and develop it into a telemedicine system that enables personalized coaching."
The study results were published on Feb. 3 in the international journal in the field of human-computer interaction (HCI), "Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies."
References
Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (2025), DOI: https://dl.acm.org/doi/10.1145/3749485