SYNCH · Project

Brain-Connected Chips That Learn and Adapt Like Living Neurons

digitalTestedTRL 5Thin data (2/5)

synch.eucoord2020.com

Imagine plugging a computer chip directly into a living brain so the two can talk to each other and learn together — like two musicians jamming, each adapting to the other in real time. That's essentially what this project built: a silicon chip with artificial synapses made from tiny memory devices called memristors, wired into the brain of a living animal. The chip doesn't just record brain signals — it forms actual connections that strengthen or weaken over time, just like real brain cells do. The result is a hybrid system where biology and silicon co-evolve, opening the door to smarter brain-computer interfaces and brain-inspired computing.

By the numbers

consortium partners

countries represented

industry partners including SMEs

38%

industry participation ratio

total project deliverables

demo deliverables including in-vivo testing

The business problem

What needed solving

Current brain-computer interfaces are one-directional and rigid — they read signals but cannot form adaptive, two-way connections with the brain. Meanwhile, conventional AI chips consume enormous power because they process information continuously rather than in energy-efficient spikes. Companies building neural implants, neuromorphic processors, or adaptive robotics need hardware that learns and evolves alongside living systems.

The solution

What was built

The project built a complete closed-loop hybrid system: a spiking neural network chip with memristor-based artificial synapses, connected bidirectionally to a living animal brain. Key deliverables include a system-on-chip with computing unit and dedicated accelerators, algorithms mapping neuronal signals to behaviour, neuromorphic synapse protocols using memristive arrays, and the full integrated system validated through in-vivo animal testing.

Audience

Who needs this

Neural implant and BCI manufacturers developing next-generation adaptive brain interfacesSemiconductor companies designing neuromorphic and brain-inspired computing chipsRobotics firms building autonomous systems that need energy-efficient adaptive processingPharmaceutical and neuroscience companies needing better tools for brain-circuit researchDefence and aerospace contractors exploring bio-hybrid computing for edge applications

Business applications

Who can put this to work

Medical Devices & Neuroprosthetics

enterprise

Target: Brain-computer interface and neural implant manufacturers

If you are a medical device company developing next-generation neural implants for paralysis, epilepsy, or movement disorders — this project developed a closed-loop brain-silicon hybrid system with neuromorphic synapses that was integrated and tested in vivo. Unlike current BCIs that only read brain signals, this technology creates two-way adaptive connections between brain and chip, potentially enabling implants that learn and improve over time. The consortium of 8 partners across 6 countries validated the full system from simulation through animal testing.

Semiconductor & Neuromorphic Computing

any

Target: Chip designers building brain-inspired processors

If you are a semiconductor company exploring neuromorphic computing architectures — this project delivered a complete system-on-chip integrating a computing unit with dedicated accelerators and memristive arrays that mimic biological synapses. The memristor-based neuromorphic synapse technology enables on-chip learning without external memory, which could dramatically cut power consumption for edge AI. With 3 industry partners and 3 SMEs in the consortium, the technology was designed with commercialization pathways in mind.

Robotics & Autonomous Systems

mid-size

Target: Companies building adaptive robots or autonomous agents

If you are a robotics company struggling with power-hungry AI processors that can't adapt in real time — this project built spiking neural network algorithms mapped to animal behaviour, running on energy-efficient neuromorphic hardware. The system processes information using event-based signaling rather than continuous computation, meaning it only uses power when something happens. The mapping algorithms between neuronal signals and behaviour could be repurposed for robots that learn from their environment the way animals do.

Frequently asked

Quick answers

What would it cost to license or integrate this neuromorphic chip technology?

The project does not publish licensing fees or unit costs. The technology was developed under an RIA (Research and Innovation Action) with 5 universities and 3 industry partners, so IP is likely shared across the consortium. Contact the coordinator at Università degli Studi di Padova to discuss licensing terms.

Can this technology scale to industrial production?

The project demonstrated a computing unit and dedicated accelerators integrated on a system-on-chip (SoC), which is a standard path toward mass production in the semiconductor industry. However, the memristive arrays and neuromorphic synapse components are still at the research-to-prototype stage, validated in controlled lab and in-vivo animal settings rather than in manufacturing environments.

Who owns the intellectual property?

IP is governed by the EU grant agreement and shared among the 8 consortium partners across 6 countries. The 3 industry partners and 3 SMEs likely hold commercial exploitation rights for their specific contributions. Specific licensing arrangements would need to be negotiated through the coordinator.

Has this been tested on living subjects?

Yes. A key deliverable confirms the full system — including the implanted interface, memristive array, and spiking neural network — was integrated and tested in vivo in an anaesthetised animal. The project also delivered mapping algorithms between neuronal signals and animal behaviour.

How does this differ from existing brain-computer interfaces?

Current BCIs mostly read brain signals one-way. SYNCH created a closed-loop system where brain and chip form bidirectional connections using neuromorphic synapses that self-reorganize — mimicking how real brain synapses undergo continuous birth, death, and adjustment. This means the silicon side adapts alongside the biological side.

What regulations would apply to products based on this?

Brain-computer interfaces fall under strict medical device regulations (EU MDR for Europe, FDA in the US). Based on available project data, the technology has been validated in animal models but would require extensive clinical trials before any human application. Regulatory pathways for neuromorphic implants are still being defined.

What technical support is available?

The consortium includes 5 universities with deep expertise in neuroscience, chip design, and memristor fabrication, plus 3 industry partners with commercialization experience. The project produced 19 deliverables including detailed algorithms, protocols, and system integration documentation that could support technology transfer.

Consortium

Who built it

The SYNCH consortium brings together 8 partners from 6 countries (Austria, Switzerland, Germany, Israel, Italy, UK), led by the University of Padova in Italy. With 5 universities and 3 industry partners (all SMEs), the 38% industry ratio signals genuine commercial interest alongside strong academic foundations. The geographic spread across major European tech hubs plus Israel gives the project access to leading neuroscience, semiconductor, and AI talent pools. For a business looking to license or co-develop this technology, the presence of 3 SME industry partners suggests there are already commercially-minded entities in the consortium who understand market needs and may be open to partnerships.

UNIVERSITA DEGLI STUDI DI PADOVACoordinator · IT
TECHNISCHE UNIVERSITAET GRAZparticipant · AT
SYNSENSE AGparticipant · CH
BAR ILAN UNIVERSITYparticipant · IL
UNIVERSITY OF SOUTHAMPTONparticipant · UK
ENGINSOFT SPAparticipant · IT
TECHNISCHE UNIVERSITAET DRESDENparticipant · DE

How to reach the team

Università degli Studi di Padova (Italy) — look for the principal investigator in the neuromorphic computing or bioengineering department

Order a report on this consortium See UNIVERSITA DEGLI STUDI DI PADOVA profile →

Next steps

Talk to the team behind this work.

Want an introduction to the SYNCH team? SciTransfer can connect you with the right researcher for your specific application — whether that's neuromorphic chips, BCI integration, or memristor technology licensing.

Order a report on this topic More in Digital & ICT