Description du projet
Une interface cerveau-ordinateur améliorée pour les personnes handicapées
Les interfaces cerveau-ordinateur (ICO) établissent une communication directe entre le cerveau et un dispositif informatique externe. À l’heure actuelle, les ICO ne peuvent recueillir des informations qu’auprès d’un nombre limité de neurones et ne peuvent donc contrôler que des mouvements simples. Le projet IoN, financé par l’UE, mise sur le développement d’ICO avancées pour aider les patients paralysés à marcher. Pour y parvenir, les chercheurs utiliseront une communication sans fil miniature et à large bande passante pour transférer les données neurales enregistrées dans plusieurs régions du cerveau. Cette résolution spatiale améliorée, ainsi que l’introduction d’un nouveau réseau de télémétrie, feront progresser la collecte et le traitement des données, offrant ainsi aux patients handicapés des ICO de meilleure qualité.
Objectif
With the increase of people suffering from various neural disorders, the need for brain-computer interfaces (BCIs) to regain sensory-motor or cognitive functions are expected to become acute in the coming decades. However, the existing BCIs can only control simple motions, e.g. grasping, and are far from realizing our vision to help paralyzed patients to walk again. This is due to the lack of a high-bandwidth wireless BCI, capable of supporting the recording from a large number of neurons with high spatial and temporal resolution, while having large spatial coverage, brain-wide.
In IoN, we target to achieve a breakthrough in the ability to transfer data from intracortical recording devices, e.g. multi-electrode arrays, by developing a transcranial telemetry system that enables the efficient transfer at high data rate from such high channel count sensors (e.g. imec’s Neuropixels with 1000 channels). Most importantly, it will also fulfil the form factor required for minimally-invasive surgery, needed to minimize the surgical risk and the complications after insertion.
Furthermore, IoN will significantly scale up brain-wide recordings, by introducing a new telemetry network that has the capacity to support 16 distributed recording nodes (enabling a total of 16,000 channels), which has never been demonstrated from any BCIs before.
To reach these challenging targets, we propose i) a novel hybrid signal propagation method to achieve a 500Mbps data rate with a 10mm^2 implant area, 20× smaller than the state-of-the-art; ii) a completely new “spike-Aloha” protocol to maximize the network capacity, supporting 166× more channels.
The technology developed in IoN will be an important transformational step to revolutionize the way neuroscientists and neurologists collect and process brain-wide neural data. By introducing this miniature, energy-efficient, and high-capacity wireless telemetry network, we want to help patients with disability to regain the quality of life.
Champ scientifique
Programme(s)
Régime de financement
ERC-COG - Consolidator GrantInstitution d’accueil
5656 AE Eindhoven
Pays-Bas