i-PROGNOSIS · Project

Smartphone-Based Early Parkinson's Detection and Personalized Digital Interventions

healthPilotedTRL 6

Imagine your phone could notice tiny changes in how you type, swipe, or hold it — changes so small you'd never spot them yourself — and flag them as early signs of Parkinson's disease. That's what this project built: software that watches how older adults naturally use their smartphones and smartwatches, then uses machine learning to catch Parkinson's symptoms years before a traditional diagnosis. Once someone is flagged, they get personalized games, sleep aids, and voice exercises right on their phone to slow things down. The team tested this with a community of over 5,000 people across six countries.

By the numbers

6.5 million

Older adults worldwide affected by Parkinson's Disease

5,000+

Older individuals targeted for the i-PROGNOSIS community

Consortium partners across 6 countries

Total project deliverables produced

55-60

Minimum age (years) of target population

The business problem

What needed solving

Parkinson's Disease affects 6.5 million older adults worldwide, but early symptoms are so subtle they go undetected for years — no blood test or routine brain scan can catch it early. By the time patients are diagnosed, significant neurological damage has already occurred, driving up care costs and reducing quality of life. Companies in digital health, senior care, and insurance lack practical tools to screen large populations for early Parkinson's signs without expensive clinical visits.

The solution

What was built

The project built a complete digital ecosystem: machine learning modules that detect early Parkinson's symptoms from natural smartphone and smartwatch use, a mobile app suite for data collection, a cloud-based intervention platform with personalized games (exercise, dietary, emotional, handwriting/voice), nocturnal sleep interventions, and assistive tools for voice enhancement and gait guidance. All components reached final versions after clinical validation and user acceptance testing.

Audience

Who needs this

Digital health companies building remote patient monitoring or neurological screening toolsSenior care operators and aging-in-place service providersHealth insurers seeking preventive screening to reduce long-term neurological care costsPharmaceutical companies running Parkinson's clinical trials needing early-stage patient identificationWearable device manufacturers looking for clinical-grade health detection algorithms

Business applications

Who can put this to work

Digital Health & Telehealth

SME

Target: Digital health companies building remote patient monitoring platforms

If you are a digital health company looking to expand your remote monitoring capabilities — this project developed machine learning modules that detect early Parkinson's symptoms from natural smartphone and smartwatch use. The system was tested across 6 countries with a target community of over 5,000 older adults and includes a full mobile app suite plus cloud-based analytics. You could integrate these behavioral sensing algorithms into your existing platform to offer neurological screening as a service.

Senior Care & Assisted Living

enterprise

Target: Senior care facility operators and aging-in-place service providers

If you are a senior care provider dealing with late-stage Parkinson's diagnoses that drive up care costs — this project built an intervention platform with personalized game suites for physical and emotional support, nocturnal interventions for sleep quality, and assistive tools for voice enhancement and gait guidance. The system targets adults over 55-60 years old and was clinically validated. These ready-made digital interventions could be deployed to your residents or home-care clients immediately.

Health Insurance & Wellness

enterprise

Target: Health insurers and corporate wellness platform providers

If you are a health insurer trying to reduce long-term neurological care costs affecting 6.5 million older adults worldwide — this project created privacy-aware behavioral analysis that runs unobtrusively on personal devices. The system uses anonymized, secure cloud archiving and distributed machine learning, meaning it can scale without compromising patient data. Early detection means earlier intervention, which epidemiological studies link to reduced risks of falls, frailty, and depression.

Frequently asked

Quick answers

What would it cost to license or integrate this technology?

The project was a publicly funded Research and Innovation Action (RIA), so core research results are generally available for licensing from the consortium. Specific licensing terms would need to be negotiated with Aristotle University of Thessaloniki as coordinator. No commercial pricing has been published in the project data.

Can this scale to thousands or millions of users?

The system was designed for large-scale deployment from the start, targeting over 5,000 older individuals during the project. It uses distributed, privacy-aware cloud architecture for data processing and machine learning, which supports scaling. The mobile apps run on standard smartphones and smartwatches, removing the need for specialized hardware.

Who owns the intellectual property?

IP is shared among the 11-partner consortium across 6 countries, led by Aristotle University of Thessaloniki in Greece. As a publicly funded EU project, exploitation plans were required. Contact the coordinator to discuss licensing for specific modules like the ML detection algorithms or the intervention platform.

Has this been tested with real patients?

Yes. The project developed final versions of all modules after refinement based on medical evaluation outcomes and user acceptance evaluation. The behavioral model was validated by correlating system measurements with clinically relevant Parkinson's risks including frailty, falls, and emotional shifts. The target community exceeded 5,000 older individuals.

How does this handle patient data privacy?

Privacy was a core design principle. The system uses anonymization, secure cloud archiving, and distributed privacy-aware machine learning. Data is collected unobtrusively from natural device use rather than clinical settings, and the architecture processes behavioral data without exposing personal identifiers.

Can this integrate with existing hospital or clinic systems?

The final integrated ecosystem includes mobile apps and a cloud-based intervention platform. Based on available project data, the system was built as a standalone ecosystem rather than an EHR plugin, but the modular architecture (separate detection modules, intervention modules, and cloud platform) suggests integration points are feasible. Two industry partners were involved in the technical development.

What specific interventions does it provide?

The intervention platform includes a Personalized Game Suite with exercise games, dietary games, emotional games, and handwriting/voice games. It also provides nocturnal interventions for sleep quality and relaxation, plus assistive tools for voice enhancement and gait rhythm guidance. All interventions were refined through clinical evaluation.

Consortium

Who built it

The 11-partner consortium spans 6 countries (Belgium, Germany, Greece, Portugal, Sweden, UK), blending 5 universities and 3 research organizations with 2 industry partners. The 18% industry ratio is modest, suggesting the technology is still closer to clinical validation than market deployment. One SME is involved. The coordinator, Aristotle University of Thessaloniki, led the effort from Greece. The mix of academic medical centers and tech partners across Northern and Southern Europe gave the project both clinical credibility and broad geographic testing conditions for the pilot community.

ARISTOTELIO PANEPISTIMIO THESSALONIKISCoordinator · EL
ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXISparticipant · EL
FACULDADE DE MOTRICIDADE HUMANAparticipant · PT
COSMOTE KINITES TILEPIKOINONIES MONOPROSOPI AEparticipant · EL
PLUX - WIRELESS BIOSIGNALS S.A.participant · PT
AGE PLATFORM EUROPEparticipant · BE
KAROLINSKA INSTITUTETparticipant · SE
KING'S COLLEGE LONDONparticipant · UK
TECHNISCHE UNIVERSITAET DRESDENparticipant · DE

How to reach the team

Aristotle University of Thessaloniki (Greece) — use SciTransfer's coordinator lookup service for direct contact details

Order a report on this consortium See ARISTOTELIO PANEPISTIMIO THESSALONIKIS profile →

Next steps

Talk to the team behind this work.

Want to explore licensing the i-PROGNOSIS detection algorithms or intervention platform for your digital health product? SciTransfer can connect you directly with the research team and help structure the collaboration.

Order a report on this topic More in Health & Biomedical