SciTransfer
DRAGON · Project

AI-Powered Pandemic Diagnosis and Precision Medicine Decision Support for Healthcare

healthTestedTRL 5
europeanlung.org

Imagine a doctor trying to figure out how serious a COVID patient's case will be — right now they're mostly guessing from experience. DRAGON built an AI system that crunches patient data across 5 countries to predict who needs intensive care and who can safely go home. It also uses molecular profiling to match patients with the right treatment faster, like a GPS for medical decisions. The whole thing runs on a privacy-safe system so hospitals can share insights without sharing raw patient data.

By the numbers
EUR 11,381,970
EU funding for development
19
consortium partners across the platform
5
countries contributing clinical data
38
project deliverables produced
7
SMEs involved in development
32%
industry participation ratio
The business problem

What needed solving

Hospitals and healthcare systems still struggle to quickly and accurately predict which patients with respiratory infections will deteriorate and which can be safely managed at home. Misdiagnosis leads to overwhelmed ICUs, wasted resources, and worse patient outcomes. There is also no easy way for hospitals across different countries to pool clinical insights without violating data privacy laws.

The solution

What was built

The project built a multi-layered AI platform: a diagnostic and prognostic nomogram for rapid risk scoring, a precision medicine module using molecular profiling, a dynamic coronavirus knowledge graph that updates with new research, and a GDPR-compliant federated machine learning system allowing multinational hospitals to train AI models without sharing raw patient data. In total, 38 deliverables were produced.

Audience

Who needs this

Hospital networks managing respiratory disease wards and ICU capacityHealth IT companies building clinical decision support softwarePharma companies running clinical trials for respiratory treatmentsPublic health agencies preparing pandemic response infrastructureMedical device companies developing point-of-care diagnostic tools
Business applications

Who can put this to work

Hospital Networks & Health Systems
enterprise
Target: Hospital groups and large healthcare providers managing emergency departments

If you are a hospital network dealing with surges in respiratory patients and uncertain triage decisions — this project developed a multifactorial diagnosis and prognosis platform that helps clinicians predict disease severity using AI. Built across 19 partner institutions in 5 countries, the system processes patient data to deliver faster, more accurate risk scores. This means better bed allocation, fewer unnecessary ICU admissions, and improved patient outcomes.

Health IT & Clinical Software
mid-size
Target: Companies building electronic health record systems or clinical decision support tools

If you are a health IT company looking to integrate AI-driven diagnostics into your platform — this project created a federated machine learning system that is GDPR-compliant and works across multinational data sources. With 38 deliverables including a dynamic knowledge graph, the technology can be embedded into existing clinical workflows. This gives your software a competitive edge with real AI-backed prognosis without requiring hospitals to share sensitive patient data.

Pharmaceutical & Biotech
enterprise
Target: Pharma companies developing antiviral treatments or respiratory therapeutics

If you are a pharma company working on treatments for respiratory infections and struggling with patient stratification in clinical trials — this project developed a precision medicine approach using molecular profiling and advanced AI. The consortium included 6 industry partners and 7 SMEs with biotech and pharma expertise. Their tools can help identify which patient subgroups respond best to specific therapies, potentially accelerating your trial timelines and reducing costs.

Frequently asked

Quick answers

What would it cost to license or deploy this technology?

The project received EUR 11,381,970 in EU funding under the Innovative Medicines Initiative (IMI2), a public-private partnership. Licensing terms would need to be negotiated with the coordinator (Universiteit Maastricht) and relevant consortium partners. Given the IMI2 structure, some outputs may have open-access components while proprietary elements remain with industry partners.

Can this scale to work across multiple hospitals or countries?

Yes — the system was specifically designed for multinational use. It was built and validated across 19 partners in 5 countries (Belgium, Switzerland, Italy, Netherlands, UK) using a federated machine learning architecture. This means it can scale without centralizing sensitive patient data, making cross-border deployment viable.

Who owns the intellectual property?

IP is shared among the 19 consortium partners according to the IMI2 grant agreement. The consortium includes 7 SMEs and 6 industry partners who likely hold commercial rights to specific components. Contact the coordinator at Universiteit Maastricht for specific licensing arrangements.

Is this compliant with healthcare data regulations?

The project explicitly built a federated machine learning system designed for GDPR-compliant use of multinational data resources. This means the AI models travel to the data rather than data traveling to a central server — a design choice specifically addressing European healthcare data regulations.

How long would integration take?

The project ran for 3.5 years (October 2020 to March 2024) and produced 38 deliverables including a dynamic knowledge graph. Based on available project data, the system went through multiple iterations of validation and optimization, suggesting mature components. Integration timelines would depend on your existing IT infrastructure and data readiness.

Does this only work for COVID or can it handle other diseases?

While developed for coronavirus, the underlying AI architecture — federated learning, molecular profiling, and decision support — is designed as a scalable platform. The objective explicitly mentions building an innovation ecosystem applicable to other coronavirus initiatives, and the core diagnostic and prognostic methods could be adapted to other respiratory or infectious diseases.

What technical support is available?

The consortium of 19 partners includes 7 universities and 3 research institutes with deep expertise in AI, clinical medicine, and data science. Post-project support would need to be arranged directly with the coordinator or relevant industry partners. The project website at the European Lung Foundation provides further contact information.

Consortium

Who built it

The DRAGON consortium is well-balanced for moving research toward commercial use. With 19 partners across 5 countries, it brings together 7 universities for scientific depth, 6 industry partners for market know-how, and 3 research institutes for technical validation. Notably, 7 of the partners are SMEs — smaller companies that typically push harder for commercialization. The 32% industry ratio is solid for a research project and signals genuine business interest. The coordinator, Universiteit Maastricht, is a well-established Dutch research university with strong ties to the healthcare sector. The inclusion of biotech, pharma, and patient organizations alongside high-tech SMEs suggests the technology was developed with real-world clinical deployment in mind, not just academic publication.

How to reach the team

Universiteit Maastricht, Netherlands — contact through the European Lung Foundation project page or the university's research office

Order a report on this consortiumSee UNIVERSITEIT MAASTRICHT profile →
Next steps

Talk to the team behind this work.

Want to explore how DRAGON's AI diagnostics or federated learning tools could fit your healthcare business? SciTransfer can arrange a direct introduction to the right consortium partner — contact us for a tailored briefing.

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 →