SciTransfer
EuroPOND · Project

AI That Predicts Brain Disease Progression to Fix Failing Clinical Trials

healthTestedTRL 4
europond.eu

Imagine you're trying to test a new Alzheimer's drug, but your patient group is so mixed — some declining fast, some slow, some with overlapping conditions — that you can't tell if the drug actually works. EuroPOND built computer models that piece together the timeline of how brain diseases like Alzheimer's and multiple sclerosis progress, even when you only have snapshots from different patients at different stages. Think of it like reconstructing an entire movie from random still frames taken from different screenings. This lets doctors and drug developers sort patients into precise subgroups so clinical trials can finally show which treatments work for which people.

By the numbers
€200B
Annual cost of Alzheimer's disease to European citizens in care and lost productivity
€4.98M
EU funding for developing disease progression models
11
Research partners across 6 countries in the consortium
37
Total project deliverables produced
The business problem

What needed solving

Clinical trials for neurological diseases like Alzheimer's fail repeatedly because patient groups are too diverse — the disease affects everyone differently, and bulk testing cannot detect which subgroups respond to treatment. Meanwhile, Alzheimer's alone costs European citizens around €200B every year, and no disease-modifying treatments are yet available. Better tools to sort patients and predict disease progression are urgently needed to unlock effective therapies.

The solution

What was built

EuroPOND built open-source software tools and statistical models that reconstruct the long-term timeline of neurological disease progression from fragmented clinical data. Their key deliverable includes a web-based application (updated from an earlier prototype) for visualizing and applying these progression models, with 37 deliverables produced in total.

Audience

Who needs this

Pharma companies running Alzheimer's or MS clinical trialsHealth-tech firms building neurological diagnostic platformsCROs (Contract Research Organizations) designing patient stratification strategiesHealth insurers modeling long-term neurological care costsNeurology clinic networks seeking better prognostic tools
Business applications

Who can put this to work

Pharmaceutical & Biotech
enterprise
Target: Pharma companies running clinical trials for neurological drugs

If you are a pharma company running clinical trials for Alzheimer's or MS treatments and facing repeated trial failures due to patient heterogeneity — this project developed data-driven disease progression models that enable fine patient stratification into subgroups. This means you can identify which patients are most likely to respond to your drug candidate, dramatically improving trial design and reducing the risk of costly bulk-response failures.

Digital Health & MedTech
mid-size
Target: Companies building diagnostic or prognostic software for neurologists

If you are a health-tech company building decision-support tools for neurologists — EuroPOND created a web-based application and open-source software tools that model long-term disease progression from routine clinical data. You could integrate these models into your platform to offer clinicians detailed predictions of how a patient's condition will evolve, improving diagnosis accuracy and care planning.

Health Insurance & Managed Care
enterprise
Target: Health insurers and managed care organizations covering neurological conditions

If you are a health insurer dealing with the enormous costs of neurological diseases — Alzheimer's alone costs European citizens around €200B every year in care and lost productivity. EuroPOND's models can help predict disease trajectories earlier and more accurately, enabling better resource allocation, earlier interventions, and more precise risk assessment for your covered populations.

Frequently asked

Quick answers

What would it cost to license or implement these disease progression models?

EuroPOND developed its tools as open-source software, which means the core models are freely available. However, integration into commercial products, customization for specific disease areas, and regulatory validation would require investment. Contact the consortium for partnership or support terms.

Can these models work at industrial scale across large patient databases?

The models were designed to reconstruct disease timelines from cross-sectional or short-term longitudinal data, which is the type of data most healthcare systems already collect at scale. The consortium of 11 partners across 6 countries tested the approach on a diverse range of datasets for both sporadic and well-phenotyped disease types.

What is the IP situation — can we build commercial products on this?

The software tools were developed as open-source, which generally allows commercial use depending on the specific license terms. For IP related to the underlying statistical methods and specific applications, you would need to check with University College London as coordinator. Based on available project data, 37 deliverables were produced, including a web-based application.

Has this been validated in a real clinical setting?

The project produced 37 deliverables over 5 years (2016-2020) and built a web-based application that was updated from an earlier prototype. The consortium included clinical and biomedical experts with access to diverse patient datasets. Based on available project data, specific clinical validation results would need to be confirmed with the coordinator.

How hard is it to integrate these models into our existing clinical trial infrastructure?

EuroPOND specifically focused on making the tools implementable as software that works with standard clinical and biomarker data. The web-based application demonstrates a ready-to-use interface. Integration complexity would depend on your data formats and IT infrastructure, but the open-source approach lowers technical barriers.

Which neurological diseases does this cover beyond Alzheimer's?

The project explicitly covered multiple sclerosis alongside Alzheimer's disease and was designed for a range of neurological conditions. The modeling approach works across disease types where progression markers can be tracked over time, making it adaptable to other neurological and potentially neurodegenerative conditions.

Is this compliant with European health data regulations?

The project was EU-funded under Horizon 2020 and involved partners from 6 European countries (Belgium, Switzerland, France, Italy, Netherlands, UK), suggesting alignment with European data standards. Specific GDPR and medical device regulation compliance for commercial deployment would need to be verified with the consortium.

Consortium

Who built it

The EuroPOND consortium brings together 11 partners from 6 countries (Belgium, Switzerland, France, Italy, Netherlands, UK), led by University College London. The team is heavily research-oriented: 4 universities and 6 research organizations, with only 1 industrial partner (9% industry ratio) and 1 SME. This composition is typical for a research-intensive project and means the technology is strong on scientific validation but will need commercial partners to bring it to market. For a business looking to adopt these tools, the low industry involvement means there is an opportunity to be an early mover in commercializing proven research, but also means you would need to invest in productization and regulatory pathways yourself.

How to reach the team

University College London (UK) — reach out to the project coordinator through CORDIS or the project website for collaboration inquiries

Order a report on this consortiumSee UNIVERSITY COLLEGE LONDON profile →
Next steps

Talk to the team behind this work.

Want a tailored brief on how EuroPOND's disease progression models could fit your clinical trial pipeline or diagnostic product? SciTransfer connects businesses with EU research teams — contact us for a matchmaking introduction.

Order a report on this topicMore in Health & Biomedical
More in Health & Biomedical
See all Health & Biomedical projects →
Automated Allergy Test Covering All Allergens in One Run, Ready to Ship
AllergyExplorer
Open →
Lab-Sized 3D Cell Microscope Replaces 12-Month Waits at Giant Synchrotron Facilities
LICENT
Open →
Blood Test Platform That Catches Rare Diseases Early Using AI-Powered Diagnostics
METAdiag
Open →