SciTransfer
HBP SGA2 · Project

Brain Research Platforms That Help Companies Build Better AI and Brain Disease Tools

healthTestedTRL 5Thin data (2/5)

Imagine trying to understand the most complex machine ever — the human brain — by building a digital twin of it. That's basically what 151 research groups across 19 countries did here, pooling EUR 88 million to create shared online platforms where scientists can store brain data, run simulations, and even test brain-inspired computer chips. Think of it like Google Maps, but for the brain: different zoom levels from single cells to entire brain regions, all connected. The end goal? Better treatments for brain diseases and smarter AI inspired by how our brains actually work.

By the numbers
EUR 88,000,000
EU contribution to this grant phase
151
consortium partners
19
countries represented
87
total deliverables produced
6
ICT research platforms built (neuroinformatics, brain simulation, HPC, medical informatics, neuromorphic computing, neurorobotics)
102
universities in the consortium
The business problem

What needed solving

Brain diseases are among the costliest health challenges globally, yet drug development success rates for neurological conditions remain extremely low — partly because we lack good computational models of how the brain actually works. Companies building AI chips, brain diagnostics, or neurological treatments are forced to work with fragmented data and disconnected tools, slowing innovation and increasing R&D costs.

The solution

What was built

Six operational ICT research platforms covering neuroinformatics, brain simulation, high-performance computing, medical informatics, neuromorphic computing, and neurorobotics. Concrete deliverables include a Medical Informatics Platform with brain disease ontologies, BIDS Apps for imaging data pipelines, a JupyterLab-based reproducible software environment integrated with brain data APIs, and an HPAC computing platform — 87 deliverables total.

Audience

Who needs this

Pharma companies with neuroscience R&D pipelines (Alzheimer's, Parkinson's, epilepsy drug targets)AI chip companies exploring neuromorphic computing architecturesHealth-tech companies building neurological diagnostic or clinical decision support toolsRobotics companies interested in brain-inspired control systemsClinical research organizations running large-scale brain imaging studies
Business applications

Who can put this to work

Pharmaceutical & Biotech
enterprise
Target: Companies developing treatments for neurological diseases (Alzheimer's, Parkinson's, epilepsy)

If you are a pharma company struggling to identify drug targets for brain diseases — this project built a Medical Informatics Platform with brain disease ontologies and imaging data pipelines that let you cross-reference clinical brain data at scale. With 87 deliverables including curated disease ontologies integrated into application services, your R&D team can search structured brain disease data instead of sifting through scattered publications.

AI & Semiconductor
enterprise
Target: Companies developing energy-efficient AI chips or neuromorphic hardware

If you are a chip maker or AI company looking for the next leap beyond conventional processors — this project advanced neuromorphic computing, building hardware that mimics how real neurons process information. The consortium included 102 universities generating brain organization data that directly informs chip architecture. If your team needs validated neuroscience data to design brain-inspired silicon, these platforms provide exactly that.

Healthcare IT & Medical Devices
mid-size
Target: Companies building diagnostic tools or clinical decision support for neurology

If you are a health-tech company building diagnostic software for neurological conditions — this project created BIDS Apps for imaging data processing pipelines and a reproducible scientific software platform integrated with brain data services. These ready-made tools can accelerate your product development instead of building brain data infrastructure from scratch.

Frequently asked

Quick answers

What would it cost to access these platforms and tools?

The HBP platforms were built as open research infrastructure, meaning many tools and datasets are publicly available at no licensing cost. However, integrating them into commercial products would likely require dedicated engineering effort and possibly collaboration agreements with consortium members. Based on available project data, specific commercial pricing models are not described.

Can these tools work at industrial scale?

The project built a High Performance Analytics and Computing (HPAC) Platform specifically designed for large-scale brain data processing, integrating resources from partners like CEA. With 151 partners across 19 countries actively using these platforms during the project, they were stress-tested at significant scale. However, commercial-grade SLAs and support would need to be arranged separately.

What is the IP situation — can we license this technology?

This was a publicly funded EUR 88 million Research and Innovation Action under Horizon 2020. Many outputs, including the JupyterLab-based software and brain disease ontologies, were released as open-access tools. Specific IP for commercial use would need to be negotiated with ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE (EPFL) as the coordinator, or with individual platform-leading partners.

Is the Medical Informatics Platform ready for clinical use?

The deliverables show multiple MIP software releases and integration of brain disease ontologies into application services like OLS-NEURO and SCAIView-Neuro. These are research-grade tools demonstrated in a scientific environment. Clinical deployment would require additional regulatory validation and certification, which was outside this project's scope.

How current is this work — is it still maintained?

The project ran from 2018 to 2020 and is now closed. However, the Human Brain Project continued under subsequent grant agreements and the EBRAINS research infrastructure (humanbrainproject.eu) carries forward many of these platforms. Check the project website for current status of specific tools.

Can these tools integrate with our existing data systems?

The project specifically built integration capabilities — the JupyterLab functionality was optimized for HBP data management solutions and data service APIs. The BIDS Apps use the Brain Imaging Data Structure, an established standard in neuroimaging. These standards-based approaches suggest reasonable integration potential with existing clinical and research data pipelines.

Consortium

Who built it

This is one of the largest EU research consortia you will encounter: 151 partners across 19 countries, backed by EUR 88 million. However, the composition tells an important story — 102 universities and 42 research organizations dominate, with only 2 industry partners and 3 SMEs (1% industry ratio). This is overwhelmingly an academic-driven infrastructure project. For a business looking to adopt these tools, the upside is that the science is deep and well-resourced; the downside is that commercial readiness, user experience, and industry-grade support may lag behind what you would expect from a product company. Your best entry point would be through EPFL (the coordinator in Switzerland) or by identifying which of the 151 partners built the specific platform relevant to your needs.

How to reach the team

ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE (EPFL), Switzerland — SciTransfer can help identify the right contact within this 151-partner consortium for your specific technology need.

Next steps

Talk to the team behind this work.

Want to know which of the 6 HBP platforms fits your product roadmap? SciTransfer can map your technical requirements to the right research team and arrange an introduction.

More in Health & Biomedical
See all Health & Biomedical projects