In an operating room, a neurosurgeon does not implant a chip so much as lay a sheet of plastic-thin film onto the brain's surface — no threads pushed into tissue, no drill bit biting deep. The film conforms to the folds of the cortex and listens. For decades the dramatic story of brain-computer interface (脑机接口) was invasive and violent: needle-like electrodes piercing grey matter. A quieter design is now reaching the clinic, and it may matter as much as the headline-grabbing implants.
Precision Neuroscience and the "Layer 7" bet
New York-based Precision Neuroscience, founded in 2021 by Benjamin Rapoport — a neurosurgeon and founding member of Neuralink — and Michael Mager, built its approach around a thin-film cortical array called the Layer 7 Cortical Interface. Each module carries 1,024 electrodes and rests on the cerebral cortex without penetrating it, inserted through a small incision in the skull. It is designed to be removable and, in its cleared form, implantable for up to 30 days.
In April 2025 the U.S. FDA granted the device 510(k) clearance — reported as the first time a company developing a next-generation wireless brain-computer interface (脑机接口) component received full FDA clearance, as opposed to a trial-only investigational exemption. The immediate use is intraoperative brain mapping: giving surgeons high-resolution functional maps before they remove an epileptic focus or a tumour, so they spare language and movement areas. The company says it began in-human testing in June 2023 and had studied the implant in dozens of patients. It has raised roughly $430 million in total, including a $250 million Series D, and announced a partnership with Medtronic to integrate Layer 7 with the StealthStation surgical navigation platform.
The strategic logic is a "middle path." Synchron's endovascular stent avoids opening the skull but offers limited electrode contact; Neuralink's penetrating threads maximise contact but invade tissue. A surface film splits the difference — more contact than a stent, far less trauma than threads — and, because it can be temporary, it cleared the regulatory bar first.
The Chinese parallel: Shenzhen's 0.01 mm film
China is not watching from the sidelines. Shenzhen-based Miling Medical (微灵医疗), founded by Li Xiaojian (李骁健), has built an ultra-flexible thin-film electrode just 0.01 mm thick — about one-tenth the width of a human hair — using MEMS and composite-nanomaterial processes. Unlike rigid grids that can damage tissue, the film hugs the cortex.
In June 2025, Miling Medical and the Shenzhen Second People's Hospital used a 128-channel flexible film array during an awake craniotomy on a 40-year-old patient with a brainstem-adjacent cavernous haemangioma. As the patient performed movement and language tasks under awake anaesthesia, the system captured high-density cortical signals in real time; the company's AI-enhanced decoder reported recognition accuracy above 97% with response times within 60 milliseconds. The same toolkit includes a portable 1,024-channel acquisition system sampling at 30 kHz per channel. Miling says its full-implant microsystem, battery-free and wireless, has passed animal tests and is moving toward clinical trials.
The parallel is striking: two continents, two thin films, the same bet that gentle surface contact is the scalable route to useful, repeatable brain-computer interface (脑机接口).
Why it matters to ordinary readers
A removable, low-trauma film lowers the risk bar for every downstream application — not just mapping during surgery, but longer-term assistive use once a chronic version is cleared. If implants stop requiring irreversible penetration of healthy tissue, far more patients and surgeons will accept them, and the manufacturing learning curve shortens. For China, domestic thin-film capability also means the supply chain for implantable brain-computer interface (脑机接口) can be local, not imported.
Honest limitations
Precision's cleared device is temporary (up to 30 days) and approved for mapping, not yet for a permanent therapeutic implant; chronic performance against Neuralink or Synchron remains unproven in head-to-head data. Miling's 97% decoding figure is from a single intraoperative demonstration on one patient, not a multicentre trial, and its chronic implant is still pre-clinical. Funding and partnership timelines for Precision come from company statements and trade press rather than audited filings. Neither firm has yet shown a permanently implanted surface film restoring function in a large patient population. The "middle path" is promising but, so far, more regulatory milestone than proven therapy.
What readers can do now
- Patients facing epilepsy or brain-tumour surgery can ask whether intraoperative brain mapping with high-density surface arrays is available at their centre — it is now an FDA-cleared option in the U.S. and in use in Chinese hospitals.
- Distinguish "temporary mapping film" from "permanent therapeutic implant": only the former is cleared today; vendors implying otherwise are ahead of the evidence.
- For those tracking the industry, watch clearance status at the FDA and China's NMPA (国家药监局) as the real signal of progress, not funding headlines.
- Investors should separate firms with a cleared, shippable component from those still in animal or early-human stages.