The Brain’s Infrastructure Era

When the Human Brain Project launched in Europe in 2013, it carried an almost impossibly ambitious promise: to bring together neuroscience, computing, medicine, and brain simulation at a scale that individual labs could never achieve alone. A decade later, the project’s most important legacy has shifted from being a single “digital brain,” to an incredible infrastructure: EBRAINS, a European digital platform now shaping how brain research is organized, shared, modeled, and translated into neurotechnology.

EBRAINS was built by the Human Brain Project (HBP) and is now being developed through EBRAINS 2.0, which runs from 2024 to 2026. Its goal is to provide researchers with open access to brain datasets, multilevel brain atlases, modeling and simulation tools, high-performance computing, robotics resources, and neuromorphic platforms. In 2024, the European Commission committed €38 million to support this next phase, with an explicit focus on neuroscience, brain medicine, and brain-inspired technologies.

Modern neuroscience has a data problem. Brain research now produces enormous amounts of information: imaging, electrophysiology, cellular data, computational models, disease datasets, and behavioral measures. But these data often remain fragmented across institutions, formats, and specialties. EBRAINS is trying to solve that less glamorous but critical problem: making brain data findable, interoperable, and usable across disciplines.

For neurotechnology, this infrastructure layer may be essential. Brain-computer interfaces, neuromodulation systems, digital twins, neurorehabilitation tools, and brain-inspired AI all depend on better ways to understand brain signals in context. EBRAINS supports that by connecting datasets with brain atlases, simulation environments, and computational workflows. Its “siibra-api,” for example, allows developers to connect EBRAINS brain atlases to applications through structured queries for brain regions, parcellations, reference templates, and regional datasets.

The platform is also trying to bridge research and industry. EBRAINS has engagement pathways for start-ups, small and medium-sized enterprises, clinical partners, and technology companies, including tool piloting, co-development, service integration, licensing, sponsorship, and national node participation. Earlier HBP industry projects included companies working on knowledge-management software, exoskeleton simulation, stroke neurorehabilitation digital twins, and brain PET imaging.

EBRAINS has not solved the mystery of the brain. The brain remains too biological, adaptive, and complex for simple digital replication. But EBRAINS is helping build something more immediately useful: the shared roads, maps, models, and computing systems that future neurotechnology will need.

For developers, clinicians, and researchers, EBRAINS represents a shift in the field. Neurotechnology is no longer just about building devices. It is about building the infrastructure that allows those devices to be scientifically grounded, clinically meaningful, and connected to the larger map of the human brain.

For investors, EBRAINS is an important signal of where neurotechnology may be heading: toward infrastructure-backed, clinically validated brain modeling. The most investable theme is the rise of digital brain twins — patient-specific computational models that combine MRI, diffusion imaging, EEG, MEG, fMRI, and other data to simulate brain network behavior.

One project to watch is Virtual Brain Twin, an EBRAINS-embedded initiative funded through a €10 million Horizon Europe grant. The project is focused on developing personalized virtual brain twins for psychiatric disorders, beginning with schizophrenia. Its stated goal is to use multiscale brain modeling, AI, high-performance computing, and clinical data to help guide medication decisions, brain stimulation strategies, and other individualized interventions.

The most relevant clinical proof point is epilepsy. EBRAINS has highlighted the EPINOV clinical trial, which is testing whether personalized virtual brain models can improve surgical planning for drug-resistant epilepsy by better identifying the epileptogenic zone — the brain network area responsible for seizure generation. EBRAINS reported in December 2025 that EPINOV is translating Virtual Brain Twin research into clinical application, and a 2025 Nature Computational Science article described a virtual epileptic patient workflow being evaluated in an ongoing prospective trial involving 356 patients.

The investor relevance is this: if virtual brain modeling proves clinically useful in epilepsy, the same infrastructure logic could extend into neuromodulation planning, psychiatric treatment optimization, neurorehabilitation, digital biomarkers, and device targeting. That creates opportunities not only for software companies, but also for neurostimulation firms, EEG/MEG platforms, imaging companies, clinical workflow vendors, and AI-enabled decision-support tools.

There are also company-level precedents. Earlier Human Brain Project industry-engagement projects included Bitbrain, which coordinated Neuro-robin, a closed-loop upper-limb neurorobot simulator for stroke neurorehabilitation; Biomax Informatics, which worked on brain knowledge-management software connected to EBRAINS data; and GEM Imaging/ONCOVISION, which worked on dedicated brain PET imaging.

The funding event to watch for may be the next phase of EBRAINS 2.0 open calls, clinical validation milestones, and industry partnerships. EBRAINS 2.0 open calls have offered €60,000 grants to external partners to integrate data, workflows, and models into the infrastructure, while EBRAINS’ current industry strategy explicitly invites start-ups, SMEs, clinical partners, and technology companies into piloting, co-development, service integration, and licensing pathways.

Investor takeaway: EBRAINS is the infrastructure layer that may help determine which neurotech companies can produce clinically grounded, validated, interoperable products. The commercial opportunity is likely to emerge around companies that can turn EBRAINS-enabled science, especially digital twins and brain simulation, into regulated, reimbursable, workflow-ready tools.

A digital brain twin is a personalized computer model of a person’s brain built from data such as MRI, EEG, MEG, fMRI, diffusion imaging, and clinical information. It is not a copy of consciousness or identity. Instead, it simulates brain structure and network activity so researchers and clinicians can test how a person’s brain may respond to disease, surgery, stimulation, or treatment.

For neurotechnology, digital brain twins could become a critical planning layer. They may help identify seizure networks in epilepsy, guide neuromodulation targets, personalize rehabilitation strategies, or predict treatment response in complex brain disorders. The function is to make brain care more precise, individualized, and testable before an intervention reaches the patient.

Clinicians are trained to be skeptical of the gap between research and real life, because we see every day that patients rarely behave as cleanly as they do in a study design. A dataset may be carefully collected, a model may perform well, and a treatment effect may look promising in a controlled research setting — but then the patient in front of us has multiple diagnoses, atypical symptoms, medication interactions, developmental differences, trauma, inconsistent access to care, or simply a biology that does not follow the expected pattern.

That is why EBRAINS feels clinically important: not because it guarantees translation, but because it points toward a future where brain research may become more integrated, comparable, and clinically usable across populations and settings. If neurotechnology is going to move from impressive signal detection to treatments that meaningfully help real patients, the world may need shared infrastructures like EBRAINS that make it easier to connect research data, clinical complexity, computational models, and treatment development in a more realistic way.