PRIMAGE · Project

Cloud Platform That Predicts Childhood Cancer Outcomes Using AI and Medical Imaging

healthTestedTRL 6

Imagine a doctor looking at a child's tumor scan and wishing they could fast-forward to see how that tumor will behave — will it grow, shrink, or respond to a specific treatment? PRIMAGE built a cloud-based tool that takes medical images, runs them through AI models and virtual tumor simulations, and gives doctors a prediction with a confidence score. It was tested on two of the most devastating childhood cancers: neuroblastoma and a lethal brain tumor called DIPG. Think of it like a weather forecast, but for how a tumor will evolve — helping doctors pick the best treatment path sooner.

By the numbers

consortium partners involved in platform development and validation

countries contributing clinical data and expertise

pediatric cancer types validated (neuroblastoma and DIPG)

industry partners including 4 SMEs for commercial pathway

project deliverables completed

The business problem

What needed solving

Pediatric oncologists face life-and-death decisions about childhood cancer treatment with limited predictive tools — they rely heavily on experience and standard protocols rather than personalized prognosis. Tumor behavior varies dramatically between patients, and current imaging analysis cannot reliably predict which treatments will work best for a specific child. Delays in identifying the right therapy path cost precious time in fast-moving pediatric cancers like neuroblastoma and DIPG.

The solution

What was built

PRIMAGE delivered a final version of its cloud-based decision support platform, combining imaging biomarkers, in-silico tumor growth simulation, and machine learning predictors for pediatric cancer clinical endpoints. The platform includes a complete data pipeline with pseudonymization, quality control, and hybrid cloud architecture (public EOSC and private clouds) across 21 deliverables.

Audience

Who needs this

Medical imaging software companies adding oncology AI featuresPharmaceutical companies running pediatric cancer clinical trialsHospital networks upgrading their oncology decision-support systemsHealth IT cloud providers serving cancer treatment centersPediatric oncology research centers seeking validated prediction tools

Business applications

Who can put this to work

Medical imaging software

mid-size

Target: Companies developing radiology or oncology decision-support tools

If you are a medical imaging software company struggling to add predictive analytics to your oncology products — this project developed a cloud-based platform with imaging biomarkers, in-silico tumor growth simulation, and machine learning predictors validated on 2 pediatric cancers across 17 partner institutions in 8 countries. The platform could be licensed or integrated to extend your existing diagnostic tools with prognostic capabilities.

Pharmaceutical and clinical trials

enterprise

Target: Pharma companies running pediatric oncology trials

If you are a pharmaceutical company running clinical trials for childhood cancer treatments and need better patient stratification — PRIMAGE built machine-learning tools trained on retrospective clinical, imaging, molecular, and genetic data from leading European pediatric oncology units. These tools predict disease-specific clinical endpoints, helping you identify which patients are most likely to benefit from experimental therapies.

Health IT and cloud services

mid-size

Target: Cloud platform providers serving hospitals

If you are a health IT provider looking to offer oncology-specific cloud services to hospitals — PRIMAGE implemented a hybrid cloud model using both public (EOSC-based) and private clouds, with built-in data pseudonymization, quality control, and secure storage. The architecture was designed for both research reuse and future commercial exploitation, giving you a ready-made blueprint for clinical cloud deployment.

Frequently asked

Quick answers

What would it cost to license or access the PRIMAGE platform?

Based on available project data, no commercial pricing has been published. The platform was built on a hybrid cloud model designed for future commercial exploitation, so licensing terms would need to be negotiated directly with the coordinator. The use of EOSC public cloud services suggests some components may be open-access.

Can this platform scale beyond the two cancers it was tested on?

The platform was validated on neuroblastoma and DIPG, but the underlying architecture — imaging biomarkers, in-silico simulation, and machine learning — is designed as a general cancer decision-support system. Scaling to other tumor types would require new training data and clinical validation for each disease. The 17-partner consortium across 8 countries provides a broad clinical data foundation.

Who owns the IP and how is it licensed?

Based on available project data, IP is likely shared among the 17 consortium partners under standard Horizon 2020 rules. The project explicitly mentions suitability for future commercial exploitation. Specific licensing terms would need to be discussed with the coordinator at Hospital Universitario La Fe in Valencia, Spain.

Has this been tested in real clinical settings?

Yes. Three of Europe's most prominent pediatric oncology units participated as consortium partners. The platform was trained and tested on retrospective clinical data including imaging, clinical, molecular, and genetic registries. Solutions for prospective data use were also validated.

How does this integrate with existing hospital IT systems?

PRIMAGE uses a hybrid cloud model combining public EOSC services and private clouds. It includes built-in data pseudonymization, extraction, structuring, quality control, and storage processes. Integration with existing hospital PACS and EHR systems would likely require customization.

Is this compliant with medical device regulations?

Based on available project data, the platform was developed as a research tool and validated in clinical research settings. Achieving CE marking or FDA clearance as a medical device would require additional regulatory steps. The project did implement data pseudonymization and security measures aligned with clinical data requirements.

What kind of ongoing support is available?

The project closed in May 2023, but the consortium includes 5 industry partners and 4 SMEs who may offer commercial support. The European Society for Paediatric Oncology was a consortium partner, providing a pathway for continued clinical adoption and support.

Consortium

Who built it

The PRIMAGE consortium of 17 partners across 8 countries is well-balanced for moving research toward market. With 8 universities providing clinical expertise and data, 3 research organizations handling scientific development, and 5 industry partners (4 of them SMEs) bringing commercial perspective, the 29% industry ratio signals genuine intent to create usable products, not just papers. The coordinator is Hospital La Fe in Valencia, Spain — a major clinical research hospital — which means the technology was built by people who actually treat patients. The presence of the European Society for Paediatric Oncology as a partner adds a powerful adoption channel across European pediatric cancer centers.

FUNDACION PARA LA INVESTIGACION DEL HOSPITAL UNIVERSITARIO LA FE DE LA COMUNIDAD VALENCIANACoordinator · ES
CHEMOTARGETS SLparticipant · ES
QUIBIM SOCIEDAD LIMITADAparticipant · ES
AKADEMIA GORNICZO-HUTNICZA IM. STANISLAWA STASZICA W KRAKOWIEparticipant · PL
KLINIKUM DER UNIVERSITAET ZU KOELNparticipant · DE
THE UNIVERSITY OF SHEFFIELDparticipant · UK
UNIVERSIDAD DE ZARAGOZAparticipant · ES
ST. ANNA KINDERKREBSFORSCHUNG GMBHparticipant · AT
UNIVERSITA DI PISAparticipant · IT
ST ANNA KINDERKREBSFORSCHUNG VEREINparticipant · AT
ALMA MATER STUDIORUM - UNIVERSITA DI BOLOGNAparticipant · IT
UNIVERSITAT KONSTANZparticipant · DE
UNIVERSITAT POLITECNICA DE VALENCIAparticipant · ES
SIOP EUROPEparticipant · BE
BC PLATFORMSparticipant · FR
ANSYS FRANCE SASparticipant · FR
MATICAL INNOVATION SLparticipant · ES

How to reach the team

Coordinator is Fundación para la Investigación del Hospital Universitario La Fe in Valencia, Spain. Use Google AI Search to find the project coordinator's direct contact.

Order a report on this consortium See FUNDACION PARA LA INVESTIGACION DEL HOSPITAL UNIVERSITARIO LA FE DE LA COMUNIDAD VALENCIANA profile →

Next steps

Talk to the team behind this work.

Want to explore licensing the PRIMAGE platform or integrating its AI imaging tools into your oncology products? SciTransfer can arrange a direct introduction to the research team.

Order a report on this topic More in Health & Biomedical