Seminal
The magazine and atlas of EU-funded researchA new publication from the web and AI team at Scott Charnley

Atlas / Netherlands

European Research Council grant · Running

AIM-HIGH AI driven data compression for high-bandwidth brain neural sensing interfaces

Coordinator
Stichting IMEC Nederland, Netherlands
Runs
1 December 2026 to 31 May 2028
Status
Running
EU funding
€150k
Total cost
—
Partners
2 organisations in 1 countries
Type
European Research Council grant
Grant agreement
101334856

What the project is doing

In the project's own words, from its summary on CORDIS.

Next-generation intracortical brain-computer interfaces (iBCIs) and large-scale neuroscience experiments require simultaneous recording from thousands of neurons at high spatial and temporal resolutions. However, current wireless systems are fundamentally limited by a bandwidth–energy bottleneck: extracellular spike recordings generate extremely large data streams, while implantable and head-mounted devices operate under strict power constraints.

As a result, existing telemetry cannot scale to high-channel-count recordings without compromising battery life and signal fidelity. This limitation restricts the precision of neural decoding and prevents BCIs and freely moving animal studies from reaching their full potential. AIM-HIGH will overcome this barrier by translating the Send-On-Delta Autoencoder (SODA) technology, developed in the ERC IoN, into a scalable wireless neural telemetry system.

SODA achieves >80× low-loss compression with <8 µV RMS distortion, preserving spike fidelity while drastically reducing data bandwidth and energy consumption. In AIM-HIGH, we will translate the SODA technology into a demonstrator, integrating the SODA ASIC with real-time decompression software in a lightweight headstage compatible with Neuropixels probes. The system will be validated in freely moving animal models to demonstrate reliable neural telemetry under realistic conditions.

These experiments will assess robustness to biological variability, including movement artefacts and electrode drift, and benchmark the system against state-of-the-art wireless headstage technologies. In parallel, the project will engage electrophysiology equipment manufacturers, neuroscience laboratories, and emerging neurotechnology companies to validate product–market fit and refine system specifications. By enabling high‑density, high‑fidelity neural telemetry, AIM‑HIGH has the potential to unlock new neuroscience experiments and accelerate the development of next‑generation BCI technology.

Who is involved

OrganisationCityCountryRole
Stichting IMEC NederlandEindhovenNetherlandsCoordinator
Erasmus Universitair Medisch Centrum RotterdamRotterdamNetherlandsParticipant

Source: CORDIS, European Commission. Data of 5 Nov 2026, reused under the Commission's reuse policy. View on CORDIS · Grant DOI