Home Sensor System Could Spot Loneliness In Older Adults

TL;DR

For listenersOffer from Amazon

Turn quiet afternoons into listening time

• Thousands of audiobooks, podcasts and originals
• Listen on your phone, tablet or Echo — also offline
• Cancel anytime

Try Audible free →Free trial for new members

As an affiliate, we earn on qualifying purchases.Hide

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 near Atlanta. The system collects routine activity, sleep and phone-use data, but researchers are still testing whether those patterns reliably correspond with loneliness assessments; it is not a diagnostic tool.

Researchers from Virginia Commonwealth University and Emory University have tested a home sensor system called HomePal in 70 independently living adults aged 65 and older, gathering routine behavior data to investigate whether it could help caregivers notice patterns associated with loneliness. The system is still being evaluated: the researchers say it is intended to flag changes for follow-up, not diagnose loneliness.

HomePal combines commercially available devices with custom sensors to monitor movement, daily activities, sleep and phone use inside a home. The prototype was installed for three months in the homes of 70 older adults living independently in the Atlanta metropolitan area. The study is published in the journal Assessment.

The devices include motion and door sensors, smart plugs connected to televisions, sleep sensors and smart speakers. Researchers also examine patterns in phone calls as a possible indicator of social connectedness. These measures describe behavior and routines; none, on its own, establishes that a person is lonely.

The team is comparing collected data with loneliness assessments given every two weeks. An AI model is being used to identify which sensed patterns are associated with participants’ reported feelings and how well those patterns correspond with assessment results. The project’s next analytical task is to test the model’s accuracy against those survey measures.

A Possible Early Signal at Home

Loneliness can be difficult for care teams to identify during a brief appointment, particularly when an older adult does not raise the subject. HomePal’s researchers are exploring whether longer-term information about changes in ordinary routines could give a caregiver a reason to check in. The intended value is earlier follow-up, not replacing conversations or clinical judgment.

If the approach proves useful, a trend in sleep, movement or communication could give family members or care professionals additional information to discuss with the person. But behavior can change for many reasons, and a sensor reading cannot explain its cause. The findings will matter only if the system can distinguish patterns that meaningfully track reported loneliness and do so without over-interpreting everyday variation.

How HomePal Collects Routine Data

The project brings together computer science and nursing researchers to compare information from home devices with participants’ own accounts. Mohamed Kamel, a computer science doctoral student at VCU and the study’s lead author, described the aim as looking for early signs in everyday behavior that might be linked to loneliness or social withdrawal. Assessments led by Chung, a former VCU School of Nursing professor, provide the comparison data.

The researchers describe a possible future product that could be installed across a home in a manner similar to a security system and produce a daily score or trend for the person or their care team. That is a longer-term goal, not a capability demonstrated as a validated commercial service in the report. The current work remains focused on analyzing the pilot data, developing models and adjusting the sensor setup.

The team also says the system could be modified to rely on sensors that process and store information locally. That option is relevant because the prototype includes smart speakers and television plugs, which raise questions about what household data is collected, who can see it and how long it is retained.

“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, VCU computer science doctoral student and lead author

Accuracy and Privacy Still Need Testing

The report does not provide a validated accuracy rate for HomePal’s AI model or show that it can reliably identify loneliness in people beyond the pilot group. The researchers are still analyzing data and comparing model outputs with biweekly assessment results. It is not yet clear which behaviors, if any, will prove consistently informative or how the model will handle changes caused by health, mobility, household circumstances or personal preference.

The study also does not establish that sensor monitoring improves health outcomes or leads to effective intervention. The proposed daily score and caregiver alerts are future aims; the report does not describe them as a tested clinical service. Further details about data retention, consent, access controls and safeguards for a potential commercial product are also not specified. The researchers say local storage and limiting what is shared may be options, but the final privacy design remains unsettled.

Model Validation and Sensor Changes

The research team plans to continue analyzing the pilot data, developing AI models and modifying HomePal’s sensors as it learns which information may be useful. A key next step is to compare model results with the participants’ loneliness assessments and determine how accurately the system reflects those reports.

Any move toward a product for homes would require further testing and decisions about how results are presented, who receives them and how residents control their information. The report does not give a release date or say that a commercial system is ready. For now, HomePal is a research prototype intended to identify possible patterns that could prompt a human follow-up.

Key Questions

What is HomePal?

HomePal is a prototype home sensing system that combines devices such as motion sensors, sleep sensors, smart speakers and television plugs to collect information about routine activity. Researchers are studying whether some patterns correspond with older adults’ reported loneliness.

Does HomePal diagnose loneliness?

No. Lead author Mohamed Kamel said the system is not intended to diagnose loneliness. The team is investigating whether certain behavioral patterns may be associated with loneliness and could warrant follow-up by a person.

How many people took part in the study?

The prototype was deployed for three months in 70 homes belonging to adults aged 65 and older who were living independently in the Atlanta metropolitan area.

How are researchers checking whether the system works?

The team is comparing sensor data with loneliness assessments collected every two weeks and is developing AI models to examine how closely sensed patterns align with participants’ reported feelings. The report does not give a final accuracy result.

What is known about privacy protections?

Researchers say HomePal could be adapted to use sensors that store data on local networks and that users could have control over what is shared with caregivers. The report does not specify a finalized privacy design for a future product.

Source: rss