TL;DR
Get everyday helpers delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Researchers from Virginia Commonwealth University and Emory University tested HomePal, a home sensor system, in the homes of 70 adults aged 65 and older living independently in metro Atlanta. The system collects data on movement, sleep, daily activity and phone use; researchers are comparing those patterns with loneliness assessments, but have not established that it can accurately detect or diagnose loneliness.
Researchers from Virginia Commonwealth University and Emory University have tested a home sensor system called HomePal in the homes of 70 adults aged 65 and older, gathering behavioral data to study whether it may help identify patterns associated with loneliness. The system’s ability to detect those patterns has not yet been established: the team is comparing the sensor data with participants’ loneliness assessments and is still developing and evaluating its models.
HomePal combines commercially available and custom sensors to passively collect information about movement, daily activities, sleep and phone use, similar to other home sensor systems for older adults. The prototype was installed for three months in the homes of older adults living independently in the Atlanta metropolitan area. Researchers plan to compare the collected data with loneliness assessments conducted every two weeks.
The sensors include motion and door detectors, smart plugs that estimate television use, sleep sensors and smart speakers. The research also examines phone-call patterns as a possible indicator of social connectedness. The team is using an AI model to investigate which behaviors, if any, are more strongly associated with loneliness and how they relate to participants’ reported feelings.
The project is described in a 2026 paper in the journal Assessment, titled “Development of an AI-Enabled Remote Loneliness Assessment System for Older Adults Using Behavioral and Speech Data From IoT Sensors and Smart Speakers.” The researchers’ stated aim is to test whether routine activity data could alert a person or caregiver to changes worth following up on—not to use the system as a diagnosis.
Could Home Patterns Prompt Earlier Check-Ins?
Loneliness can be difficult for health professionals to identify during a clinic appointment, particularly when a person does not raise it directly. A system that tracks changes in everyday routines could, if validated, offer caregivers another way to notice possible concerns and start a conversation. The study addresses a practical challenge: loneliness is a personal experience, while the sensors record behavior that may or may not reflect how someone feels.
For families and care teams, a future tool might show changes over time rather than rely on a single visit or self-report. But the research has not shown that HomePal can reliably identify loneliness, improve care or produce better health outcomes. Its value will depend on whether the patterns prove meaningful, whether alerts are useful rather than misleading, and whether older adults are comfortable with the monitoring.
As an affiliate, we earn on qualifying purchases.
How HomePal Links Sensors and Surveys
The project brings together computer science and nursing researchers. The group is comparing sensor readings with biweekly loneliness assessments, giving it a way to test whether recorded routines correspond with participants’ own reports. That comparison is still underway, and the supplied report does not give accuracy results or identify a validated loneliness score.
The researchers envision a possible future product that could be installed in a home much like a security system and provide a daily score or trend for an older adult or caregiver. That is a longer-term goal, not a current capability shown by the study. The present work centers on data collection, model development and deciding which sensors may be useful.
“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 VCU computer science doctoral student
Accuracy, Privacy and Use Remain Open
The report does not provide results showing how accurately HomePal can identify loneliness, which specific signals are most informative, or how often the system might raise an alert. It also does not establish that a change in movement, sleep, television viewing or phone use is caused by loneliness; many factors could affect those routines. The researchers say they are still analyzing data and building models.
Privacy and consent are also unresolved questions for any future deployment. The team says HomePal could be adjusted to use sensors and storage on a local network, and that data-sharing choices could be limited. The report does not specify a final commercial design, data-retention policy or how a future system would handle inaccurate alerts. No commercial availability date is given.
Researchers Continue Model Testing
The team plans to continue analyzing the data gathered during the three-month home deployment, comparing it with the participants’ recurring assessments and refining its AI models and sensor setup. The next important evidence will be whether those models can identify consistent associations and how well they perform against the survey-based measures.
Until the researchers report validation results, HomePal remains a research prototype, not a proven screening or diagnostic tool. Any later move toward a product would also require decisions about privacy, access to data and how caregivers should respond to a flagged pattern.
Key Questions
What is HomePal?
HomePal is a research prototype that uses household sensors and connected devices to collect information about older adults’ movement, activity, sleep and phone use. Researchers are studying whether those patterns relate to reported loneliness.
Has HomePal been shown to detect loneliness accurately?
Not yet, based on the reported findings. The research team is still comparing sensor data with biweekly loneliness assessments and evaluating its AI models. The report gives no accuracy results.
Who took part in the study?
The prototype was deployed for three months in the homes of 70 adults aged 65 and older who lived independently in metro Atlanta.
Does HomePal diagnose loneliness?
No. Lead author Mohamed Kamel said the team is looking for behavior patterns that may be related to loneliness and could merit follow-up. The system is not presented as a diagnostic tool.
How could the researchers address privacy?
The team says the system could be adapted to use sensors and data storage on local networks, with options to control what is shared with caregivers. The final privacy design has not been specified, as the project is still in development.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
