Neural chip decodes interactions between brain regions

What happened: EPFL researchers developed LogSync, a low-power implantable neural system-on-chip that records activity across brain regions, extracts network-level biomarkers, and uses a machine-learning algorithm to classify cognitive and emotional states and trigger stimulation. Key numbers: Power consumption under 200 Why it matters: Most existing deep brain stimulation systems are open loop or rely on signals from a single brain region, which the researchers describe as a major limitation for psychiatric disorders where relevant information is distributed across interacting networks.

EPFL researchers developed LogSync, a low-power implantable neural system-on-chip that records activity across brain regions, extracts network-level biomarkers, and uses a machine-learning algorithm to classify cognitive and emotional states and trigger stimulation.

Most existing deep brain stimulation systems are open loop or rely on signals from a single brain region, which the researchers describe as a major limitation for psychiatric disorders where relevant information is distributed across interacting networks.