FrailSafe · Project

Wearable Sensors and AI That Predict Elderly Frailty Before Hospitalization

healthTestedTRL 5

Imagine your aging parent seems fine one day and ends up in the hospital the next — frailty sneaks up like that. FrailSafe built a system of wearable sensors and smart software that continuously watches for subtle warning signs — changes in walking, speech, sleep, even how someone plays a game on a tablet. It feeds all that data into a digital patient model that can predict when someone is sliding toward a fall, a hospital visit, or worse. The goal is to catch frailty early enough to actually reverse it with personalized exercises, games, and lifestyle changes.

By the numbers

consortium partners across the project

European countries represented in the consortium

SME partners involved in development

56%

industry ratio in consortium

total deliverables produced

The business problem

What needed solving

Frailty in elderly populations leads to sudden hospitalizations, increased care costs, and poor outcomes — yet it often goes undetected until a crisis hits. Current clinical assessments are infrequent snapshots that miss the gradual decline happening between visits. Care providers and insurers need continuous, early warning systems that can predict deterioration and trigger preventive action before costly emergency interventions become necessary.

The solution

What was built

FrailSafe produced a Decision Support System prototype that predicts clinical state changes using a digital patient model, signal processing algorithms that extract frailty indicators from wearable sensors despite noise and data gaps, a linguistic analysis tool (LingTester) that detects cognitive decline through speech patterns, offline multi-parametric data analysis tools, and augmented reality serious games for anti-frailty interventions — totaling 40 deliverables across the project.

Audience

Who needs this

Residential elderly care and assisted living facility operatorsHealth insurance companies covering aging populationsWearable health device manufacturers seeking clinical applicationsTelehealth and remote patient monitoring platform providersMunicipal and regional public health authorities managing elderly care programs

Business applications

Who can put this to work

Elderly Care & Assisted Living

mid-size

Target: Residential care facility operators and home care providers

If you are an elderly care provider dealing with unpredictable patient deterioration and emergency hospitalizations — FrailSafe developed a Decision Support System prototype that uses wearable sensor data to predict clinical decline. The system fuses data from physical, cognitive, psychological, and social monitoring to flag at-risk residents before a crisis happens, letting your staff intervene with personalized programs instead of reacting to emergencies.

Health Insurance & Managed Care

enterprise

Target: Health insurers covering elderly populations

If you are a health insurer struggling with rising hospitalization costs for frail elderly members — FrailSafe created signal processing algorithms that extract frailty indicators from wearable sensors and a prediction engine that models clinical state changes. This could let you shift from paying for expensive emergency care to funding cheaper preventive interventions, using continuous remote monitoring tested across 6 European countries.

Medical Devices & Wearables

any

Target: Wearable health technology manufacturers

If you are a wearable device company looking to move beyond fitness tracking into clinical-grade elderly monitoring — FrailSafe developed validated signal processing algorithms that extract frailty-related indicators from standard wearable sensors despite noise and incomplete data. Their methods cover physical activity, cognitive function via linguistic analysis, and behavioral patterns — giving your hardware a clinically meaningful software layer with 40 deliverables of R&D behind it.

Frequently asked

Quick answers

What would it cost to license or adopt this technology?

Based on available project data, no pricing or licensing terms are disclosed. The project was a publicly funded Research and Innovation Action (RIA) with 9 consortium partners. Commercial terms would need to be negotiated with the coordinator (University of Patras) and the relevant technology-owning partners.

Can this scale to monitor thousands of elderly patients simultaneously?

The system was designed for continuous unobtrusive monitoring using wearable sensors, with signal processing algorithms built to handle noise and incomplete data at scale. However, based on the deliverables, the system was demonstrated in research settings, not yet validated at large-scale commercial deployment with thousands of simultaneous users.

Who owns the intellectual property and how can I access it?

The project was funded under Horizon 2020 as an RIA, meaning IP typically stays with the partners who generated it. With 5 SME partners and 5 industry organizations in the consortium, IP is likely distributed across multiple entities. Contact the coordinator at University of Patras (Greece) to discuss specific licensing arrangements.

Does this comply with medical device regulations in the EU?

The project focused on research and prototype development, producing a Decision Support System and sensor algorithms. Based on available deliverables, there is no evidence of CE marking or Medical Device Regulation (MDR) certification. Any commercial deployment would require regulatory clearance as a medical device or clinical decision support tool.

How long would integration take with our existing care management systems?

The project produced standalone prototypes including a Decision Support System, offline analysis tools, and signal processing modules. Based on the deliverable descriptions, these were built as research prototypes, so integration with commercial care management platforms would require additional engineering work — likely a dedicated development phase.

What evidence exists that this actually works with real patients?

The project ran from 2016 to 2019 across 6 countries with clinical partners. Deliverables include both preliminary and final versions of the Decision Support System, tested with patient data and physiological measurements. The LingTester component was tested in both passive and active modes with real linguistic data from patients.

Consortium

Who built it

The FrailSafe consortium brings together 9 partners from 6 European countries (Belgium, Cyprus, Greece, Spain, France, Italy), coordinated by the University of Patras in Greece. What stands out for business adoption is the strong industry presence: 5 of the 9 partners are SMEs and the overall industry ratio is 56%, meaning more than half the consortium has commercial DNA. This mix of 1 university, 2 research organizations, and 5 industry players suggests the technology was developed with real-world application in mind, not just academic publication. The multi-country spread also means the system was likely tested across different healthcare contexts, which matters for any company considering European-wide deployment.

PANEPISTIMIO PATRONCoordinator · EL
INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALEparticipant · FR
ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXISparticipant · EL
BRAINSTORM MULTIMEDIA SLparticipant · ES
HYPERTECH (CHAIPERTEK) ANONYMOS VIOMICHANIKI EMPORIKI ETAIREIA PLIROFORIKIS KAI NEON TECHNOLOGIONparticipant · EL
AGE PLATFORM EUROPEparticipant · BE
SMARTEX SRLparticipant · IT
GRUPPO SIGLA SRLparticipant · IT

How to reach the team

Coordinator is University of Patras (PANEPISTIMIO PATRON), Greece. SciTransfer can help locate the right contact person and facilitate an introduction.

Order a report on this consortium See PANEPISTIMIO PATRON profile →

Next steps

Talk to the team behind this work.

Want to explore how FrailSafe's frailty prediction technology could fit your elderly care or health insurance product? SciTransfer connects you directly with the research team — contact us for a tailored one-page brief and introduction.

Order a report on this topic More in Health & Biomedical