Independent living, safely watched over.
ElderSafe Home places discreet IoT sensors around the home and uses a CNN-LSTM model to recognise daily activities in real time, alerting family and caregivers to falls or unusual inactivity.
The problem
Most older people want to live at home, safely.
Falls and emergencies often happen when no one is around. Cameras feel intrusive, and wearables are often forgotten. Ambient sensors offer a respectful middle ground.
people die from falls each year worldwide; adults over 60 are most affected.
activity recognition accuracy of our model.
average latency, fast enough for real-time alerts.
Sources: WHO falls fact sheet; results from the founding research paper (ICBATS, 2025).
How it works
How it works
A clear path from data to decision.
Install
Small IoT sensors are placed in key rooms.
Sense
Signals are fused in real time and cleaned of noise.
Recognise
A CNN-LSTM model classifies daily activities.
Alert
Family and caregivers are notified of risks immediately.
Product
What makes it different.
Every feature is grounded in the founding research.
No cameras
Ambient sensors protect privacy and dignity.
Real time
150 ms average latency.
Accurate
92.5% accuracy, 91.0% precision, 90.2% recall.
Instant alerts
Notifications to family and caregivers when something is wrong.
Who it's for
Who it's for.
Families
Peace of mind for relatives living alone.
Home-care providers
Remote oversight between visits.
Assisted-living facilities
Scalable monitoring across residents.
The research
Built on peer-reviewed research.
- End-to-end real-time human activity recognition in smart homes.
- Hybrid CNN-LSTM model with real-time sensor fusion.
- 92.5% accuracy, 91.0% precision, 90.2% recall, 90.6% F1-score.
- Average latency of 150 ms, suitable for real-time applications.
Radhi, A. D., et al. (2025). Real-Time Human Activity Recognition in Smart Homes Using IoT Sensors and Deep Learning Models. ICBATS 2025, IEEE.
Team
The team behind it.
Founded in 2025 by a student team from two faculties.
Faculty of Science (Computer Science)
IoT, deep learning and software.
Faculty of Nursing
Elderly care needs, safety protocols and user testing.
FAQ
Questions we hear often.
Does it use cameras?
No. It relies on ambient IoT sensors to protect privacy.
Who receives alerts?
Family members and caregivers you choose.
Is this a finished commercial product?
It is a research-based student startup moving from validated prototype toward pilots with real users.
Can we partner or run a pilot?
Yes. We are looking for organisations willing to test the solution in real conditions and share feedback.
Bring ElderSafe Home to your organisation.
We are onboarding pilot partners now. Read the research and get in touch.