BCIwiki — Brain-computer interface news, research and industry database

Why it matters The gain comes from where the fusion happens rather than from a bigger model, which suggests the ceiling on EEG-to-image retrieval is partly a ranking-architecture problem, and it is cheap enough for other groups to test before peer review settles the claim.

Why it matters The first thin-film electrode product to win long-duration implantation clearance, it maps a regulatory route that Chinese players on the same flexible-electrode track — NeuroXess and StairMed — can benchmark their registration strategies against.

Why it matters Moving from demonstrations to several hours of paid computer work is closer to real employment, but reproducibility will depend on standardized metrics and complete clinical data.

Why it matters Harmonizing heterogeneous montages at the signal level — instead of retraining per device — is what lets motor imagery decoders transfer across datasets and hardware, a prerequisite for moving beyond a single lab setup.

Why it matters Chinese BCI labs and companies have each used their own data formats, to the point where EEG recorded by one group could not be read by another's algorithms and cross-institution transfers took days of conversion. These are the first standards to address that nationally. Two qualifiers matter. They are recommended standards, so nothing is illegal about ignoring them and their real force depends on industry uptake and whether procurement documents cite them. And neither is a privacy or neuroethics instrument, despite being reported that way. The one privacy-adjacent provision is a naming rule in GB/T 47127 requiring subject directories to carry a sub- prefix and barring real names and ID numbers, which is de-identification housekeeping inside a format spec rather than a rights framework.

Why it matters The power budget, not electrode resolution, is what keeps high-bandwidth implants tethered or battery-limited, so a backscatter-plus-wireless-charging design that removes in-implant transceivers directly targets the constraint blocking fully implantable, high-data-rate BCIs.

Why it matters This preprint matters because BCI preprocessing has been manual, expert-dependent, and poorly reproducible, a bottleneck for laboratories without dedicated expertise. By automating the planning and execution of preprocessing pipelines across six signal types with a two-phase LLM agent, EasyBCI could make auditable, reproducible preprocessing more accessible, and its design principles may apply to AI agents in other scientific domains.

Why it matters A longitudinal, multi-stage stroke dataset — including an unusual 'sixth-finger' paradigm — fills the data gap that has kept motor-imagery BCIs and neuroplasticity studies short of real rehabilitation-trajectory evidence.

Why it matters China's first high-resolution visual BCI clinical implant, using a device the hospital team has developed since 2014 — together with Xiangya's IMIE case it marks homegrown visual BCI entering clinical validation

Why it matters The interesting number is not the effect size but the dose: three weeks at five sessions a week on top of standard care is a schedule a rehabilitation ward can actually staff, which is where most BCI stroke protocols fail long before the decoding does.