As populations age and the desire to remain at home deepens, a team at Nottingham Trent University has answered a quiet but urgent question: how do you watch over someone without making them feel watched? By weaving together radar, thermal sensing, and appliance monitoring under a single AI intelligence, researchers have sketched a vision of care that is present without being intrusive — a kind of technological companionship that speaks only when something is wrong.
AI System Monitors Older Adults' Health at Home Without Cameras or Wearables
Homes that quietly monitor residents in the background
Mark
So the system never actually sees the person's face or body in detail?
Mimi
Right. The thermal imaging is deliberately pixelated and low-resolution. It knows posture—sitting, standing, lying down—but not who is doing it or what they look like.
Luke
But the radar and the smart plugs together—don't those create a pretty detailed behavioral profile? You know when someone gets up, when they use the stove, when they're in bed.
Mimi
They do. That's the point. The system learns what that person's normal day looks like, so it can spot when something changes.
Mark
What counts as an emergency? How does the AI know the difference between someone taking a nap and someone who has fallen?
Mimi
The thermal sensor can detect posture changes—if someone is lying on the floor instead of on a bed, for instance. And the radar picks up whether they're moving. If they're immobile in an unusual position for too long, that triggers an alert.
Luke
But that's still an inference, right? The system doesn't know for certain what happened. It just knows something looks different.
Mimi
Exactly. That's why it alerts a caregiver or family member to check. The AI is a first responder, not a diagnostician.
Mark
How long does it take to learn someone's normal pattern?
Mimi
The study doesn't specify, but the idea is that over time—weeks, probably—the system builds a baseline for that individual.
Luke
And what if someone's routine is genuinely erratic? What if they're an artist who works at odd hours, or someone with insomnia?
Mimi
The system would need to adapt to that. The researchers tested it in controlled environments, so real-world variability is still an open question.
Mark
Does it work for people with dementia or cognitive decline?
Mimi
The research doesn't address that. It's designed to detect acute events and gradual changes in mobility and routine, but someone with dementia might have unpredictable patterns that the system would struggle to interpret.
Luke
So this is really for people who are cognitively intact but physically vulnerable—people who might fall or have a stroke, but whose daily patterns are relatively stable.
Mimi
That seems to be the target population, yes.
Le Pouls
- Millions of older adults living alone face a silent gamble every day — a fall, a stroke, or a cardiac event with no one nearby to respond.
- Existing solutions force an uncomfortable trade-off: cameras feel like surveillance, wearables get forgotten or go uncharged, and residential care means surrendering independence.
- Nottingham Trent University researchers have layered radar, low-resolution thermal imaging, and smart plug data into a system that sees daily life without ever seeing a face.
- An AI engine learns each person's individual rhythms — when they boil the kettle, how they move through rooms — and raises an alert the moment something breaks the pattern.
- Early testing across multiple mock room layouts showed the three technologies outperform any single sensor alone, catching both sudden emergencies and the slow drift of gradual decline.
- The system is now pointed toward a future where aging in place is safer and longer, and where caregivers are extended into homes they cannot always physically enter.
As populations age and the desire to remain at home deepens, a team at Nottingham Trent University has answered a quiet but urgent question: how do you watch over someone without making them feel watched? By weaving together radar, thermal sensing, and appliance monitoring under a single AI intelligence, researchers have sketched a vision of care that is present without being intrusive — a kind of technological companionship that speaks only when something is wrong.
Researchers at Nottingham Trent University, working alongside Integrated System Technologies Ltd, have developed a home monitoring system designed to protect older adults living alone — without cameras, without wearables, and without surrendering privacy. The system combines millimetre-wave radar to track movement, deliberately pixelated thermal sensors to recognize posture, and smart plugs wired into household appliances to follow daily routines. An AI engine binds these three streams together, building a portrait of what normal looks like for each individual and alerting caregivers when something disrupts it.
The motivation is both personal and demographic. Older people increasingly want to age in place rather than enter residential care, but solitary living carries genuine risk. A fall in the bathroom or a stroke in the kitchen may go unnoticed for hours. Traditional monitoring tools tend to feel like surveillance or require daily maintenance — burdens that erode the very independence they are meant to protect. This system aims at a middle ground: protection that stays in the background.
The radar sensors detect movement through walls and furniture at low cost. The thermal sensors are intentionally coarse — capable of distinguishing sitting from standing or lying down, but unable to resolve faces or fine detail. The smart plugs register when the kettle boils, the stove activates, or the computer wakes. Together, they offer what the team describes as a complete picture of how someone is managing day to day.
Testing in a mock home across six room configurations confirmed that the combined system outperforms any single sensor working alone. Beyond catching acute emergencies, the AI can also flag subtler signals — reduced mobility, disrupted sleep, skipped meals — that may indicate early decline worth a caregiver's attention. When an anomaly is detected, alerts go to family members or care professionals, who can then decide whether to call or visit.
Lead researcher Dr. Yangang Xing framed the work as a direct response to shifting demographics, while industry partner Dr. Geoff Archenhold described the goal as an ambient assisted living environment — a home that watches quietly and speaks only when it must. The research, published in the journal Sensors, represents one carefully considered answer to a question that will only grow more pressing: how to keep people safe at home without turning home into an institution.
Researchers at Nottingham Trent University have built a monitoring system that watches over older adults living alone at home—detecting falls, strokes, and other health crises—without a single camera pointed at them or a device strapped to their wrist. The system, developed in partnership with Integrated System Technologies Ltd, layers three separate sensing technologies: millimetre-wave radar to catch movement, thermal imaging to recognize posture, and smart plugs wired into household appliances to track daily routines. An artificial intelligence engine stitches the data together, learning what normal looks like for each person over time, then alerting caregivers or family when something breaks the pattern.
The appeal is straightforward. Many older people want to stay in their own homes as long as possible rather than move into residential care facilities. But living alone carries real risk—a fall in the bathroom, a stroke in the kitchen, a heart attack at night. Traditional solutions feel invasive: cameras in bedrooms, wearable devices that require charging and remembering to wear. This system aims at a middle ground, offering protection without the surveillance.
The radar sensors are placed at chosen spots around the home, detecting when someone moves from room to room. They are inexpensive and work through walls and furniture. The thermal sensors are deliberately kept low-resolution and highly pixelated—they can tell if a person is sitting, standing, walking, or lying down, but they cannot capture facial features or fine details of movement. The smart plugs monitor when appliances turn on and off: the kettle, the stove, the computer. Together, these three data streams create what the research team calls a complete picture of how someone is managing day to day.
During testing, the researchers set up a mock home with six different room layouts, including bedrooms and living areas with various furniture arrangements. They performed a series of activities and measured how well the system could identify movement and posture. The results showed that the three technologies worked better together than separately. When the AI had access to all three data sources, it caught more than when it relied on any single sensor alone.
Dr. Yangang Xing, the lead researcher from NTU's School of Architecture, Design and the Built Environment, framed the work as a response to a demographic shift. Older adults are living longer and often prefer independence. The system is designed to support that preference while catching emergencies early. Beyond detecting acute events like falls or cardiac episodes, the AI can also flag gradual changes—a person moving less, sleeping differently, skipping meals, breaking their usual patterns. These shifts might signal the beginning of decline that warrants a check-in from a caregiver or a visit to a doctor.
When the system detects something unusual, it sends an alert to care professionals and family members, who can then reach out remotely or visit in person to assess what is happening. The goal is not to replace human judgment but to extend it—to give caregivers eyes and ears in a home they cannot always be in, without turning that home into a surveillance space.
Dr. Geoff Archenhold, chief executive of Integrated System Technologies Ltd, described the vision as an ambient assisted living environment—homes that quietly monitor residents in the background and speak up only when something seems wrong. He suggested the technology could help older people remain independent longer, giving them confidence to move through their days knowing that their wellbeing is being watched over without their privacy being invaded.
The research was published in the journal Sensors under the title "Privacy-Preserving Ambient Sensing for Activities of Daily Living: Multimodal Radar-Thermal Human Activity Recognition and Smart Plug Appliance Recognition." The team included researchers from both Nottingham Trent University and Integrated System Technologies Ltd. The work represents one approach to a growing challenge: how to keep older people safe at home without making their homes feel like institutions.
Citations marquantes
By combining these three technologies, we can establish a complete picture of how someone is coping at home. As well as detecting emergencies, it could provide an early warning if a person's mobility or daily routine begins to deteriorate.— Dr. Yangang Xing, Nottingham Trent University
The technology could ultimately allow older people to stay in their own homes for longer, giving them confidence to go about their daily business while feeling secure in the knowledge that their health and wellbeing is monitored without infringing on their privacy.— Dr. Geoff Archenhold, Integrated System Technologies Ltd
Envie de l'histoire complète ?
Lire l'original sur News-Medical ↗
Technology