TL;DR
Researchers from Virginia Commonwealth University and Emory University tested HomePal, a home sensor system, in the homes of 70 adults aged 65 and older for three months. They are analyzing whether data on activity, sleep and phone use correlate with loneliness assessments; the system is still under development and is not a diagnostic tool.
Researchers from Virginia Commonwealth University and Emory University have tested a home sensor system called HomePal in the residences of 70 independently living adults aged 65 and older, aiming to identify everyday behavior patterns that may be associated with loneliness. The three-month study, published in the journal Assessment, is intended to compare sensor data with participants’ loneliness assessments; the researchers say the system is not designed to diagnose loneliness.
Turn your wind-down time into listening time
- Thousands of audiobooks, podcasts and originals
- Listen on your phone, tablet or Echo — also offline
- Cancel anytime
HomePal combines commercially available and custom devices to passively collect information about movement, daily activities, sleep and phone use. The setup included motion and door sensors, smart plugs attached to televisions, sleep sensors and smart speakers. Researchers also examined patterns in phone calls as one possible measure of social connection.
The team is comparing those data with loneliness assessments conducted every two weeks. The research group is using an AI model to examine which observed behaviors are more strongly associated with participants’ reported feelings. The next research task is to test how closely the model’s results match the assessment responses, rather than treating any single behavior as proof of loneliness.
The study is described in a 2026 paper by Mohamed Kamel and co-authors, “Development of an AI-Enabled Remote Loneliness Assessment System for Older Adults Using Behavioral and Speech Data From IoT Sensors and Smart Speakers.” The reported pilot involved 70 adults in the Atlanta metro area; researchers say they are still analyzing the information and adjusting the system.
A Possible Earlier Signal for Care Teams
Loneliness can be difficult for care teams to recognize during occasional clinic visits, particularly if an older adult does not raise the subject. A system that tracks changes in ordinary routines could, if future testing supports its use, give an older adult or caregiver a reason to check in sooner. That may make it easier to discuss support and care options, but the study has not shown that HomePal improves health outcomes or prevents loneliness.
The approach also shifts some attention from self-reported feelings alone to patterns collected at home over time. That could complement conversations and assessments, not replace them. The researchers’ stated aim is to flag changes for follow-up, leaving interpretation and decisions to people rather than the sensor system.
Products Worth Considering
Levoit Eleva 300X Air Purifier for Home Bedroom
Fitbit Versa 4 Health and Fitness Smartwatch with GPS, Heart Rate and Sleep Tracking, Alexa, Water Resistant, Black
Oral-B iO Series 3 Rechargeable Electric Toothbrush for Adults, Black, with 2 Brush Heads and Travel Case
As an affiliate, we earn on qualifying purchases.
How HomePal Connects Sensors and Surveys
The research addresses a recognition challenge: loneliness is a personal experience, while sensors capture actions that may have several explanations. For example, changes in sleep, television viewing or calls could reflect many aspects of a person’s routine and do not establish that the person feels lonely. HomePal’s design therefore pairs behavioral information with regular participant assessments.
Researchers say they hope a future version could resemble a home security system in how it is installed and could provide a daily score or trend for a participant or care team. That is a proposed direction, not a product currently shown to reliably detect loneliness. The study’s findings are being used to develop and evaluate the model, and the account does not provide performance results such as its accuracy or error rate.
“The main idea is to see whether everyday behavior at home can give early signs that an older adult may be experiencing loneliness or social withdrawal.”
— Mohamed Kamel, lead author and computer science doctoral student at Virginia Commonwealth University
Accuracy, Privacy and Clinical Use
The report does not give the model’s accuracy, sensitivity or false-alert rate, and the researchers are still comparing its outputs with assessment results. It is not yet clear which sensor patterns, if any, will reliably correspond with loneliness across different people or homes. The pilot’s reported size and duration also do not establish how the system would perform in broader settings.
Privacy is another open question. The researchers say HomePal could be adapted to use sensors that process and store information on local networks, and that data sharing could be limited. However, the account does not specify a finalized product’s data-retention rules, security protections, consent process or exact information available to caregivers. The project has not established that HomePal is ready for clinical use or commercial sale.
Researchers Continue Model Testing
The team plans to continue analyzing the pilot data, developing its AI models and adjusting the sensor setup as it learns which measures are informative. A central next step is to compare model results with the biweekly loneliness assessments and determine how accurately the system identifies patterns that merit human follow-up.
Researchers have described a potential future product that could report a daily score or trend to an older adult, family member or caregiver. The source does not give a release date or say that such a product is available. Further results will be needed to establish performance, appropriate use and privacy safeguards before the system’s practical role can be judged.
Key Questions
What is HomePal?
HomePal is a research prototype that combines home sensors and connected devices to collect information about activity, sleep and other routines that researchers are studying for possible links with loneliness.
How many people took part in the study?
The researchers deployed the system for three months in the homes of 70 independently living adults aged 65 and older in the Atlanta metropolitan area.
Can HomePal diagnose loneliness?
No. Lead author Mohamed Kamel said the project is intended to identify behavior patterns that may warrant follow-up, not to diagnose loneliness directly.
Has the system been shown to work accurately?
The report does not provide accuracy results. Researchers are still comparing the model’s findings with participants’ biweekly loneliness assessments.
What happens to the data?
The researchers say the system could be adapted to store data locally and give patients control over what is shared with caregivers. Specific safeguards and final data rules for any future product have not been reported.
Source: rss
